WO2024045064A1 - 播放内容的监测方法、主机、***、装置、介质及设备 - Google Patents
播放内容的监测方法、主机、***、装置、介质及设备 Download PDFInfo
- Publication number
- WO2024045064A1 WO2024045064A1 PCT/CN2022/116276 CN2022116276W WO2024045064A1 WO 2024045064 A1 WO2024045064 A1 WO 2024045064A1 CN 2022116276 W CN2022116276 W CN 2022116276W WO 2024045064 A1 WO2024045064 A1 WO 2024045064A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- content
- identification code
- monitored
- content identification
- intermediate information
- Prior art date
Links
- 238000000034 method Methods 0.000 title claims abstract description 105
- 238000012544 monitoring process Methods 0.000 title claims abstract description 99
- 238000012795 verification Methods 0.000 claims abstract description 132
- 230000002159 abnormal effect Effects 0.000 claims abstract description 73
- 238000001514 detection method Methods 0.000 claims description 70
- 230000005856 abnormality Effects 0.000 claims description 39
- 230000008569 process Effects 0.000 claims description 28
- 238000006243 chemical reaction Methods 0.000 claims description 25
- 238000012549 training Methods 0.000 claims description 22
- 238000013528 artificial neural network Methods 0.000 claims description 9
- 239000000284 extract Substances 0.000 claims description 7
- 238000004590 computer program Methods 0.000 claims description 6
- 238000012806 monitoring device Methods 0.000 claims description 5
- 238000004891 communication Methods 0.000 abstract description 7
- 238000010586 diagram Methods 0.000 description 14
- 238000005516 engineering process Methods 0.000 description 11
- 238000007689 inspection Methods 0.000 description 11
- 230000007246 mechanism Effects 0.000 description 8
- 238000004458 analytical method Methods 0.000 description 5
- 239000003814 drug Substances 0.000 description 4
- 229940079593 drug Drugs 0.000 description 4
- 230000010365 information processing Effects 0.000 description 4
- 238000010801 machine learning Methods 0.000 description 4
- 238000012015 optical character recognition Methods 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000013145 classification model Methods 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 101100217298 Mus musculus Aspm gene Proteins 0.000 description 1
- 230000006978 adaptation Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000005336 cracking Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 238000004880 explosion Methods 0.000 description 1
- 230000007274 generation of a signal involved in cell-cell signaling Effects 0.000 description 1
- 239000013067 intermediate product Substances 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 231100000614 poison Toxicity 0.000 description 1
- 230000007096 poisonous effect Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 238000001454 recorded image Methods 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000011144 upstream manufacturing Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
- H04N21/23—Processing of content or additional data; Elementary server operations; Server middleware
- H04N21/24—Monitoring of processes or resources, e.g. monitoring of server load, available bandwidth, upstream requests
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
- H04N21/442—Monitoring of processes or resources, e.g. detecting the failure of a recording device, monitoring the downstream bandwidth, the number of times a movie has been viewed, the storage space available from the internal hard disk
Definitions
- the present disclosure relates to the field of communication technology, and in particular, to a monitoring method, host, system, device, medium and equipment for playing content.
- Media resource management systems are widely used in transportation, finance, shopping malls, supermarkets and other scenarios.
- Media resource management systems usually include display devices, which are used for playback Specified media content, such as advertisements, film and television dramas, promotional information, etc.
- the present disclosure provides a method for monitoring playback content.
- the method includes:
- determining whether the content to be monitored is compliant based on the first content identification code includes:
- the first-level verification fails, perform a second-level verification on the display page screenshot, and the second-level verification is used to determine the abnormal category of the content to be monitored;
- secondary verification is performed on the display page screenshot, including:
- performing screen anomaly detection on the screenshot of the display page to determine whether there is a first type of abnormal screen and/or a second type of abnormal screen in the content to be monitored includes:
- the first detection model is obtained by training the first neural network using a plurality of images carrying screen abnormalities as training samples
- the second detection model is obtained by using a plurality of images carrying abnormal content as training samples.
- the second neural network is trained.
- a second-level verification is performed on the display page screenshot, including:
- the method when it is determined that the content to be monitored is non-compliant, the method further includes at least one of the following:
- the method before the step of obtaining a screenshot of the display page from the display device during the current monitoring period, the method further includes:
- the second content identification code is used as a comparison with the first content identification code to determine whether the content to be monitored is compliant.
- determining whether the content to be monitored is compliant based on the first content identification code includes:
- determining whether the content to be monitored is compliant based on the first content identification code includes:
- the second content identification code and the first content identification code are respectively analyzed to obtain first intermediate information corresponding to the second content identification code and the first content identification code respectively; wherein, the first The intermediate information is the intermediate information in the process of converting the identity coding information corresponding to the content into the content identification code;
- the first intermediate information is a summary of the identity encoding information of the played content.
- the second content identification code further includes a hidden area, the hidden area is used to carry second intermediate information corresponding to the identity encoding information, and the second content identification code is based on the second content identification code and the The steps of determining whether the content to be monitored is compliant with the first intermediate information respectively corresponding to the first content identification codes include:
- Information wherein the second intermediate information is produced before the first intermediate information;
- determining whether the content to be monitored is compliant based on the second intermediate information corresponding to the second content identification code and the first content identification code includes:
- the method further includes:
- an alarm signal of a corresponding level is generated; wherein the content attributes include at least one of an identification attribute of the display device and a playback period attribute.
- generating a second content identification code of the predetermined content based on the identity coding information includes:
- the first intermediate information is decimal grayscale information, and values at different positions in the first intermediate information correspond to check points at different positions.
- the step of iteratively performing multiple encodings on the identity encoding information to obtain intermediate information under each encoding includes;
- an information hiding area is set on the predetermined transparent image, and the information hiding area is used to carry second intermediate information corresponding to the identity encoding information.
- the method further includes:
- the second intermediate information is produced earlier than the first intermediate information, and the second intermediate information is binary information;
- the gray value of the information hidden area is updated.
- the step of generating identity coding information corresponding to the predetermined content includes:
- the multi-dimensional content attributes include one or more of the name of the predetermined content, the playback period, and the device identification of the display device that plays the predetermined content;
- the multi-dimensional content attributes are respectively encoded in hexadecimal format to obtain the identity encoding information.
- the present disclosure also provides a playback control host, which is used to execute the method for monitoring playback content.
- the present disclosure also provides a broadcast monitoring system, including: a content publishing server, at least one broadcast control host connected to the content publishing server, and a display device connected to each of the broadcast control hosts;
- the content publishing server is used to send the content to be played to the corresponding broadcast control host;
- the playback control host is used to send the content to be played to the display device, and to execute the method for monitoring the playback content
- the display device is used to play the content to be played.
- the present disclosure also provides a computer-readable storage medium, the computer program stored therein causes the processor to execute the method for monitoring playback content.
- the present disclosure also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor.
- the processor executes, the described playback of content is realized. Monitoring methods.
- the present disclosure also provides a monitoring device for playing content, which device includes:
- a screenshot acquisition module configured to obtain a screenshot of a display page from a display device during the current monitoring period, where the screenshot of the display page is obtained by taking a screenshot of the content to be monitored currently played by the display device;
- An identification code acquisition module configured to extract a first content identification code from the display page screenshot, where the first content identification code is used to identify the played content
- a verification module configured to determine whether the content to be monitored is compliant based on the first content identification code.
- the technical solution of the embodiment of the present disclosure can obtain a screenshot of the display page from the display device during the current monitoring period, extract the first content identification code from the screenshot of the display page, and then determine whether the content to be monitored is appropriate based on the first content identification code. regulation.
- the screenshot of the display page is obtained by taking a screenshot of the display page of the display device, and the first content identification code is used to identify the currently played content to be monitored, it is possible to determine what kind of content is currently played by the display device based on the content identification code. content, so that it can be determined whether the content to be monitored is compliant based on the first content identification code.
- the first content identification code can be used to determine whether the content to be monitored is content designated for playback, and whether an abnormality occurs on the playback page of the content to be monitored, thereby improving the accuracy of inspection of the played content.
- Figure 1 schematically shows a schematic diagram of a communication environment in which a method for monitoring playback content of the present disclosure is applied;
- Figure 2 schematically shows a step flow chart of a method for monitoring playback content according to the present disclosure
- Figure 3 schematically shows a flow chart of yet another method for monitoring playback content of the present disclosure
- Figure 4 schematically shows an overall flow chart of a method for monitoring playback content of the present disclosure
- Figure 5 schematically shows a principle diagram of converting text-type identity encoding information into image-type content identification codes according to the present disclosure
- Figure 6 schematically shows a schematic diagram of encoding identity encoding information in an example of the present disclosure
- Figure 7 schematically shows a complete flow diagram of the first-level verification of the present disclosure
- Figure 8 schematically shows a complete flow diagram of the secondary verification of the present disclosure
- Figure 9 schematically shows the framework structure of a monitoring and broadcasting system of the present disclosure
- Figure 10 schematically shows a structural diagram of a content monitoring device of the present disclosure.
- Media Asset Management (MAM) system is based on advanced technologies such as multimedia, network, database and digital storage to digitize various media and content (such as video/audio materials, text files, charts, etc.)
- the overall solution for storage, management and application including all aspects of digital media collection, cataloging, management, transmission and coding conversion. It mainly meets the requirements of media resource owners to collect, save, search, edit, and publish various information, provides users of media resources with convenient methods to access content, realizes efficient management of media resources, and greatly improves the efficiency of media resources. value.
- the business pain point of the media resource management system is: after the content is released, it is necessary to monitor whether the display device is broadcasting according to the scheduled plan. This is because the display device may be invaded by a third party, causing the display device to be played by a third party involving pornography. Illegal content such as poisonous or politically sensitive content; of course, in some cases, it is also necessary to promptly monitor whether the display device has a black screen or a blurred screen, whether the content being played should be played on the current display device, and whether the playback time period of the content being played is correct, etc.
- broadcast monitoring is generally carried out through manual spot patrols, which is costly and inefficient, and cannot handle emergencies in a timely manner.
- each display device can be equipped with a monitoring camera, and the monitoring camera can be pointed at the display screen of the display device to record the video, so that the screen content can be automatically analyzed through AI and other technologies based on the recorded image. This is used to monitor the content being played on the display device.
- this method has the following problems:
- each display device needs to be equipped with a surveillance camera, which not only increases the cost.
- a surveillance camera you also need to follow specific standards such as height and angle to install the camera, and install the camera-related structures, wiring and other projects. It is restricted at many customer sites and difficult to implement in project implementation.
- the present disclosure proposes a method for monitoring playback content.
- the core concept of this method is:
- This content identification code generally uniquely represents a content to be played, and can be displayed on the display page and played along with the content. In this way, when monitoring the playback content, Take a screenshot of the playback page of the display device, and then extract the content identification code from the screenshot, so that based on the content identification code, it can be determined whether the content being played by the display device is compliant.
- FIG. 1 a schematic diagram of the communication environment of a media resource management system of the present disclosure is shown. As shown in Figure 1, it includes: a content publishing platform, a content publishing server, a plurality of broadcast control hosts and each broadcast control host.
- the content publishing server is connected to multiple broadcast control hosts, and one broadcast control host can be connected to one or more display devices of the same type, or multiple display devices of different types.
- the serial port/HDMI can be used as the downstream path for data between the broadcast control host and the display device to send the content to be played
- the network port/USB can be used as the upstream path for data to receive screenshots of the display page uploaded by the display device.
- operators can log in to the content publishing platform to publish the content to be played, specify the display device corresponding to the content to be played, and then click Publish.
- the content publishing platform can then publish the content to be played and the content to be played.
- the identification of the corresponding display device is also sent to the broadcast control host through the content publishing server, and then the broadcast control host sends the content to be played to the corresponding display device.
- the content publishing platform when users log in to the content publishing platform to publish content to be played, they can also specify the playback period of the content to be played.
- the content publishing server can also send the playback period to the broadcaster.
- the control host instructs the display device to play the content to be played in the corresponding playback period.
- the content of the present disclosure may refer to multimedia data, such as advertisements, TV programs, etc.
- a method for monitoring playback content of the present disclosure is introduced.
- FIG 2 a step flow chart of a method for monitoring playback content of the present disclosure is shown.
- the monitoring method of playing content in the example can be executed by the broadcast control host shown in Figure 1.
- it can also be executed by the content publishing server, or can be executed by other terminals in the media resource management system.
- the other terminal device only needs to communicate with the display device. This disclosure does not specifically limit the execution mechanism of the method for monitoring playback content.
- Step S201 During the current monitoring period, obtain a display page screenshot from the display device.
- the display page screenshot is obtained by taking a screenshot of the display page of the display device.
- the current monitoring period may refer to the period during which the display device plays content. In practice, it may refer to the working period of the display device. For example, if the display device needs to play content between 10:00 and 18:00 of a day, then the current monitoring period may refer to the period during which the display device plays content. The monitoring period can be from 10:00 to 18:00. In this case, screenshots of the display page can be obtained from the display device at preset intervals for timely content monitoring.
- the display device may refer to the period during which each content is played during the working period of the display device.
- the display device needs to play content from 10:00 to 18:00 of the day, where content 1 is 10:00-12:00, content 2 is played at 12:10-13:00, then the current monitoring period can refer to the period of the currently playing content 1 or the period of the currently playing content 2.
- Screenshots of the display page can be obtained during the playback period corresponding to each content to monitor the playback of each content.
- a screenshot of the display page can be obtained from the display device during the current monitoring period.
- a screenshot instruction can be sent to the display device at any time during the current monitoring period, and the display device can respond to the screenshot instruction. Take a screenshot of the display page to obtain a screenshot of the display page.
- the screenshot of the display page can be returned to the broadcast control host or server that sent the screenshot instruction.
- the display page of the screenshot may be the display screen of the display device, thereby retaining all content currently displayed on the display screen.
- the display page of the screenshot may also be a playback window in which content is currently playing on the display screen. In practice, you can take a screenshot of the full screen or the playback window according to the actual situation.
- Step S202 Extract a first content identification code from the screenshot of the display page.
- the first content identification code is used to identify the content to be monitored currently played by the display device.
- content that needs to be played on the display device can have a corresponding content identification code generated in advance.
- the content identification code can be used to identify the content identification, name, category, playback period and other attributes of the played content.
- the content identification code of a content can be a globally unique identification code to uniquely represent a content.
- the content identification code can be a text type identification code, a string type identification code, or it can be similar to a QR code. identification code.
- the content identification code can also be displayed on the display page of the display device. In this way, when taking a screenshot of the display page, you can capture the screen of the played content. And intercept the content identification code. Therefore, when a screenshot of the display page is obtained, the first content identification code can be extracted from the screenshot of the display page.
- the display device when it plays content, it can display the content identification code corresponding to the content to a fixed position on the display page.
- the fixed position can be the endpoint position of the display screen or a fixed position in the playback window. position, so that whether it is a screenshot of the display page or a screenshot of the playback window of the content, it is guaranteed to have the first content identification code in the screenshot of the display page.
- the image of the first content identification code can be extracted from the image area corresponding to the fixed position in the display page screenshot, and the first content identification code can be The image is parsed to obtain the first content identification code.
- the first content identification code is a text type or string type identification code
- OCR optical character recognition, optical character recognition
- the image of the first content identification code can be identified to obtain the first content identification code represented by it.
- Step S203 Determine whether the content to be monitored is compliant based on the first content identification code.
- the content identified by the first content identification code can be determined based on the first content identification code.
- the identified content can be the content to be monitored currently played by the display device, so that the content identified by the first content identification code can be determined based on the first content identification code. code to determine whether the content to be monitored is compliant.
- the identified first content identification code is incomplete, it indicates that the display of the display page is abnormal, or it may indicate that the display page of the display device may be invaded by a third party. Therefore, based on the first content identification code, Whether the code is incomplete, determine whether the content to be monitored is compliant; if it is incomplete, it is not compliant; if it is complete, further monitoring will be performed.
- the identified first content identification code is complete, it can be detected whether the first content identification code is garbled.
- Garbled characters may mean that the first content identification code includes information that should not appear. In practice, when This situation may occur when the display device is invaded by a third party. Therefore, as long as the first content identification code is garbled, it can be determined that the content to be monitored is not compliant.
- compliance may mean that the content to be monitored complies with the internal rules of the platform.
- the specific rules can be set by the platform's needs. For example, the playback period of the content to be monitored conforms to the scheduled playback period, and the content to be monitored is the designated playback content. Or the content to be monitored does not include scenes that are prohibited from being played, etc.
- the content attributes of the content to be monitored can also be parsed from the first content identification code.
- the content attributes may include the name of the content, content identification, playback period and display device as described in the above embodiment. logo, etc. In this way, it can be determined whether the content to be monitored is compliant based on the parsed content attributes.
- the media resource management system can further inspect the correctness of the content being played on the display device to ensure that the display device plays the specified content that should be played within a playback period.
- the technical solution of this embodiment on the one hand, you only need to take a screenshot of the current display page of the display device, and the compliance of the playback content can be verified based on the content identification code extracted from the screenshot of the display page, thereby eliminating the need for manual inspection. Inspection reduces inspection costs and improves inspection efficiency.
- the first content identification code can be used to determine whether the content to be monitored is content designated for playback, and whether an abnormality occurs on the playback page of the content to be monitored, thereby improving the accuracy of inspection of the played content.
- the first content identification code since it is based on the first content identification code in the screenshot of the display page, it is determined whether the content to be monitored is compliant.
- the first content identification code generally has a small amount of data and consumes less computing resources for analysis and analysis, thus achieving Achieve fast and accurate monitoring of content with a small amount of information. It avoids the need to parse and analyze images related to the content to be monitored in the screenshots of the display page, which consumes a lot of computing resources, takes a long time to analyze, and relies heavily on algorithm accuracy (such as machine learning, which relies on the number of samples and neural network design). The problem.
- the display device when it is determined that the content to be monitored is not compliant, can also be turned off to prohibit the display device from playing content; or an alarm signal is sent to indicate that the content to be monitored is not compliant, Or turn off the display device and send an alarm signal at the same time.
- the compliance can also be in compliance with the internal rules of the platform, and the specific rules can be set by the platform requirements.
- the display device can be forcibly turned off. For example, a power-off signal can be sent to the display device to cut off the power of the display device, thereby forcing the display to The device is powered off so that the display device is turned off and illegal content to be monitored cannot be played.
- an alarm signal can also be sent to the host computer and the corresponding user.
- the host computer can be the content publishing platform described in Figure 1, and the corresponding user can be an inspector who inspects the display device. user.
- the alarm signal can be alarm information, which can carry a screenshot of the display page and the device identification of the display device, thereby indicating which display device is playing non-compliant content and the screen of the non-compliant content, thereby helping the user based on
- the alarm information needs to be further processed for monitoring content.
- the method for monitoring the playback content of the present disclosure may include technical means related to the monitoring mechanism, as well as information processing related methods.
- technical means the monitoring mechanism is mainly reflected in the monitoring process to determine whether the content to be monitored is compliant, while the information processing is mainly reflected in how to generate the content identification code and how to monitor based on the content identification code.
- this disclosure requires monitoring whether the display device is used by a third party to play pornographic, drug-related or other sensitive content and other illegal content; and it also needs to promptly monitor whether the display device has a black screen or a blurry screen, and whether the content being played should be displayed on the current display device. Playback, and whether the playback time period of the content being played is correct, etc. Therefore, the monitoring task is divided into the anomaly detection stage of the content to be monitored, and the compliance determination stage after the abnormality is detected in the content to be monitored.
- the anomaly detection stage may be based on the detection of the first content identification code, with the purpose of determining whether the content has display abnormalities and whether the content being played is specified content.
- the compliance determination stage may be based on the abnormality detection stage based on screenshots of the display page.
- the cause of the exception can be investigated. Specifically, it can be determined whether the cause of the exception is a screen display failure or a third-party intrusion.
- FIG. 3 a schematic flow chart of yet another method for monitoring playback content of the present disclosure is shown. As shown in Figure 3, the process may include the following:
- S31 During the current monitoring period, obtain a screenshot of the display page from the display device, and extract the first content identification code from the screenshot of the display page.
- the first content identification code is used to identify the currently played content to be monitored; the display page screenshot is obtained by taking a screenshot of the current screen of the display device.
- a screenshot of the display page can be obtained from the display device during the playback period of each content, so that different contents can be monitored separately during the playback period corresponding to the different content, so that the playback period of the content can be monitored. , showing whether the device is playing the content correctly.
- the start playback time and the end playback time of each content can be stored in advance, so that a screenshot of the display page can be obtained from the display device during the playback period between the start playback time and the end playback time of each content.
- the display page screenshot may be obtained by taking a screenshot of the current screen of the display device, which includes a picture of the content to be monitored and also includes the first content identification code.
- S32 Perform first-level verification based on the first content identification code. Wherein, the first-level verification is used to determine whether there is an abnormality in the content to be monitored.
- the first-level verification can be understood as preliminary verification, and its main task is to initially verify whether the content to be monitored is compliant.
- the first-level verification process can be as described in the above-mentioned step S203, that is, it can be determined whether the first content identification code is incomplete. If it is incomplete, it can be determined that there is an abnormality in the content to be monitored; if it is not incomplete, You can continue to determine whether the first content identification code is garbled. If garbled characters appear, the first-level verification fails, indicating that there is an abnormality in the content to be monitored. If no garbled characters appear, you can continue to parse out the content attributes identified by the first content identification code.
- the content to be monitored determines whether the content to be monitored is during the playback period to be monitored, and specify the content that the display device should play. If it is determined that the content to be monitored is not the content that should be played during the playback period to be monitored, or the If the identification of the display device corresponding to the content to be monitored is not the current display device, it can be determined that the content to be monitored is abnormal and the first-level verification fails; otherwise, it is compliant and the first-level verification passes.
- the first-level verification if the first-level verification is passed, it means that the content to be monitored is compliant, and the monitoring can be ended without entering S33; on the contrary, if the first-level verification is not passed, it means that the content to be monitored is compliant. In the case of non-compliance, continue to enter S33 and perform secondary verification to identify the cause of the anomaly before finally determining whether the content to be monitored is compliant.
- S33 Perform secondary verification on the screenshot of the display page.
- the secondary verification is used to determine the abnormal category of the content to be monitored, and based on the results of the secondary verification, determine whether the content to be monitored is compliant.
- the second-level verification is mainly used to check the exception category where the content to be monitored is abnormal.
- This exception category can be understood as the cause of the exception in the content to be monitored.
- the screenshot of the displayed page is It is obtained by taking a screenshot of the current screen of the display device. It not only carries the display screen and the first content identification code of the content to be monitored, but also carries other display screens on the current screen. It is rich in information, so abnormal detection can be carried out based on the screenshot of the display page. Attribution analysis, and then based on the analyzed anomaly categories, determine whether the content to be monitored is compliant.
- the second-level verification is the above-mentioned compliance determination stage.
- the abnormal reasons may include screen display failure and third-party intrusion, and this Both reasons can be derived based on taking screenshots of display pages that carry rich information.
- screen anomaly detection can be performed on the display page screenshots to determine whether the content to be monitored contains a first type of abnormal screen and/or a second type of abnormal screen; the first type is the type of screen abnormality, The second type is a type containing content exceptions.
- the first type is the type of screen abnormality.
- the screen abnormality may refer to: a display abnormality occurs on the display screen of the display device, which is a software and hardware failure of the display device. Specifically, a screenshot of the display page can be detected by Whether there is a black screen and/or blurred screen, determine whether the abnormality of the content to be monitored is attributed to the screen abnormality.
- the abnormality of the content to be monitored is attributed to the screen anomaly; if there is a blurred screen in the displayed page screenshot, it can be determined that the number of blurred screens is greater than the preset number. In the case of , it is determined that the abnormality of the content to be monitored is attributed to the screen abnormality; if there are black screens and blurred screens in the screenshots of the display page, it is determined that the abnormality of the content to be monitored is attributed to the screen abnormality. Among them, if the number of blurred pictures is not greater than the preset number, the abnormal attribution of the content to be monitored can be determined based on the results of the second type of abnormal picture detection.
- the second type is a type of abnormal content.
- it can be determined based on the images in the image area except the first content identification code in the screenshot of the displayed page whether there is any abnormal content in the content to be monitored, where the content Abnormalities may refer to images containing pornographic, drug-related or other sensitive content.
- content anomalies can be detected based on machine learning methods.
- the display device can be forcibly closed and an alarm signal can be sent to indicate enhanced media resource management. System security protection against third-party intrusion.
- machine learning can be used to obtain Monitoring results.
- the method of machine learning can refer to using training samples to train a corresponding detection model in advance, so as to analyze the display page screenshots through the detection model, thereby obtaining whether the content to be monitored contains the first type and the second type of abnormal pictures. the result of.
- a screenshot of the display page can be input to the first detection model, and based on the result output by the first detection model, it can be determined whether the content to be monitored contains the first type of abnormal screen. And/or, input the screenshot of the display page into the second detection model, and determine whether there is a second type of abnormal screen in the content to be monitored based on the results output by the second detection model.
- the first detection model is obtained by training the first neural network using multiple images carrying screen abnormalities as training samples
- the second detection model is obtained by training the second neural network using multiple images carrying abnormal content as training samples. obtained through training.
- a screenshot of the display page may be input to the first detection model to determine whether the content to be monitored contains the first type of abnormal screen.
- the results output by the first detection model may be two probability values, respectively corresponding to the category in which the abnormal picture of the first type exists and the category in which the abnormal picture of the first type does not exist.
- the category corresponding to the larger probability value can be selected as the detection result.
- the probability values are 0.8 and 0.2 respectively. 0.8 is the category corresponding to the presence of the first type of abnormal picture, then it can be determined that the detection result is the content to be monitored. There is a first type of abnormal picture.
- the output result of the second detection model may also be two probability values, respectively corresponding to the category where the second type of abnormal picture exists and the category where the second type of abnormal picture does not exist.
- the category corresponding to the larger probability value can be selected as the detection result.
- the probability values are 0.7 and 0.4 respectively. 0.7 is the category corresponding to the presence of the second type of abnormal picture, then it can be determined that the detection result is the content to be monitored. There is a second type of abnormal screen.
- the output result of the second detection model determines whether there is a second type of abnormal picture based on the content to be monitored. If there is a second type of abnormal picture, it is determined that the content to be monitored is not compliant. If there is no abnormal picture of the second type, Then it is determined that the content to be monitored is compliant but the display device is determined to be abnormal.
- the training samples of the first detection model can be sample images including black screens and blurred screens
- the second detection model can collect images containing pornographic, drug-related or other sensitive content as training samples based on the abnormal content to be detected, where , use the training samples for training, and the process of obtaining the first detection model and the second detection model can adopt the content of relevant technologies, which will not be described again here.
- the first detection model can be a commonly used classification model, for example, it can be any one of the models: VGG19, Resnet512, and inceptionV3; the second detection model can also be a commonly used classification model, such as any of VGG19, Resnet512, and inceptionV3.
- the first detection model and the second detection model can be models of the same category, such as VGG19, or they can be different types of models.
- the generalization performance of the model can be relied upon to improve the accuracy of the content to be monitored. accuracy of regulatory monitoring.
- the method for monitoring playback content of the present disclosure can be executed by the broadcast control host.
- the first-level verification can be performed by the broadcast control host; the second-level verification is then Can be performed by the content publishing server.
- the broadcast control host can send a screenshot of the display page to the content publishing server, so that the content publishing server can perform a second-level verification. ; After that, the broadcast control host can receive the second-level verification result returned by the content publishing server, and based on the second-level verification result, determine whether the content to be monitored is compliant.
- the process of the content publishing server performing the secondary verification may refer to the process described in S33 above, and will not be described again here.
- the content publishing server obtains the result of the second-level verification, it can send the result of the second-level verification to the broadcast control host, and then the broadcast control host determines whether the content to be monitored is compliant based on the result of the second-level verification.
- the result of the second-level verification indicates that there is a second type of abnormal picture, and it is determined that the content to be monitored is not compliant, and the broadcast control host forcibly closes the display device; if it indicates that there is no abnormal picture of the second type, it is determined that the content to be monitored is not compliant. Monitor content compliance.
- the content publishing server after the content publishing server obtains the results of the secondary verification, it can also directly output information on whether the content to be monitored is compliant without notifying the broadcast control host of the results. When it is not compliant, an instruction to close the display is directly given to the broadcast control host, thereby forcibly shutting down the display device.
- first-level verification and second-level verification when the first-level verification based on the first content identification code fails, the second-level verification based on the screenshot of the displayed page will be performed. Verification, so that the first-level verification and the second-level verification are based on information of different amounts of data. Therefore, there is no need to conduct global (display page screenshots and first content identification codes) every time the content to be monitored is monitored. Analysis, instead, first perform monitoring based on the first content identification code with a small amount of data. If it fails, then analyze based on screenshots of the display page with a large amount of data. This can improve monitoring efficiency and accuracy while avoiding long-term monitoring. A high proportion of computing resources is consumed, thereby improving the utilization of computing resources.
- the first-level verification is performed by the broadcast control host and the second-level verification is performed by the content publishing server, the first-level verification with a smaller amount of data is performed on the broadcast control host.
- the broadcast control host and the display The device is directly connected. There are generally multiple broadcast control hosts in the media resource management system. One broadcast control host is connected to one or more display devices. It is an edge device and can use multiple edge terminals to decentrally monitor the display devices. Efficient monitoring of live content, and due to decentralized detection, the pressure of network concurrency is also avoided.
- the content publishing server will perform a second-level verification with a large amount of data.
- the content publishing server can reasonably utilize the advantages of the content publishing server and the accuracy of the computing resources to determine compliance; in this way, a hierarchical response is adopted.
- This strategy helps the media resource management system to respond quickly, while reducing the concurrency pressure on the network, reducing the computing pressure on the content publishing server, and improving the monitoring efficiency of the media resource management system.
- the content identification code can uniquely identify the content being played, when a screenshot of the display page is obtained from the display device, the screenshot of the display page includes a picture of the content being played.
- the content identification code can be used to identify the name and playback period of the content being played. , the display device being played and other properties.
- different content may be played in different playback periods.
- the content to be monitored can also be specified to be played in the current playback period.
- the second content identification code corresponding to the scheduled content is used as a reference, that is, as a comparison with the first content identification code, to verify whether the content to be monitored is the scheduled content designated to be played in the current playback period.
- the identity encoding information of the content to be played can be generated before each content to be played is sent to the display device.
- the identity encoding information can include the name of the content to be played, the playback period, the encoding of the display device being played, and other information.
- the second content identification code of each content to be played can be obtained based on the identity encoding information, and the second content identification code can be stored locally, so that subsequent determination can be made based on the first content identification code and the second content identification code. Whether the content to be monitored is compliant.
- FIG. 4 an overall flow diagram of a method for monitoring playback content of the present disclosure is shown. As shown in Figure 4, it specifically includes the following processes:
- a second content identification code corresponding to the content to be played can be generated, where the content to be played can be understood as scheduled content played by the designated display device.
- identity coding information corresponding to the predetermined content may be generated; and then, based on the identity coding information, a second content identification code of the predetermined content may be generated.
- the identity encoding information can be a code with a specified number of digits obtained by encoding the name of the content, content identification, playback period, identification of the display device, etc., which can be understood as the identity ID of the predetermined content.
- the identity encoding information can be associated with predetermined content and a display device that plays the predetermined content.
- the identity encoding information can have a specified number of digits to reflect the multi-dimensional content attributes of the predetermined content; in one example, the identity encoding information can be the name of the content, the content identification, the playback period and the display device. The identification, etc. are obtained after encoding in hexadecimal, which means that the identity encoding information can be in hexadecimal. Of course, in some other examples, the identity encoding information can also be in hexadecimal. Information; the details can be determined based on the type and number of content attributes that are actually to be encoded. For example, the identity encoding information can be 26-bit hexadecimal identity information. The specific encoding process and content identification code generation process will be detailed in the subsequent section 2.1.2, and will not be described in detail here.
- the identity encoding information can be converted into the second content identification code.
- the conversion process can be to encrypt the identity encoding information and then obtain the second content identification code, or the second content identification code can be to encrypt the identity encoding information and then convert it.
- the obtained QR code can be obtained as long as the identity encoding information can be converted to obtain the second content identification code.
- the specific format of the second content identification code is not limited here. For details, please refer to 2.1.2. section details.
- the subscription content to be played can be sent to the corresponding display device for playback.
- the second content identification code can also be sent to the corresponding display device. Sent to the display device, so that the display device displays the second content identification code on the display page when playing the predetermined content.
- the second content identification code can be displayed at a fixed position on the screen. As mentioned above, it can be displayed at the upper left corner, upper right corner, lower left corner or lower right corner of the screen. It can also be displayed in the upper left corner, upper right corner, lower left corner or lower right corner of the playback window.
- the second content identification code can be displayed at the four corners of the screen.
- S43 Use the stored second content identification code as a comparison with the first content identification code to determine whether the content to be monitored is compliant.
- the second identification code corresponding to the reservation content can be stored in association with the playback period corresponding to the reservation content. In this way, in a playback period, when the first content identification code corresponding to the content to be monitored currently played by the display device is obtained, , the stored second content identification code corresponding to the playback period can be used as a comparison with the first content identification code to determine whether the content to be monitored is compliant.
- a second content identification code corresponding to the playback period to be monitored can be obtained, where the second content identification code is consistent with the playback period to be monitored.
- the second content identification code when using the second content identification code as a reference to determine whether the content to be monitored is compliant, in an optional example, it may be determined directly based on the first content identification code and the second content identification code whether the content to be monitored is compliant. , that is, it is not necessary to perform subsequent second-level verification, and determine whether the content to be monitored is compliant based on the first-level verification results alone. During specific implementation, you can compare whether the first content identification code and the second content identification code are consistent.
- the display device plays the predetermined content during the playback period to be monitored, and the content to be monitored is compliant; if they are inconsistent, , it means that the display device may not play the scheduled content during the playback period to be monitored, and the content to be monitored is non-compliant.
- a monitoring mechanism that combines primary verification and secondary verification can still be used, that is, when comparing the first content identification code and the second content identification code for inconsistency,
- secondary verification can be performed based on screenshots of the display page to determine whether the content to be monitored is compliant.
- monitoring may first be performed based on the first intermediate information corresponding to the first content identification code and the second content identification code. Specifically, based on the corresponding first intermediate information, the third content identification code is determined. Whether the first content identification code is consistent with the second content identification code, and based on the result of the consistency, it is determined whether the content to be monitored is compliant.
- the second content identification code and the first content identification code can be analyzed separately to obtain the first intermediate information corresponding to the second content identification code and the first content identification code respectively; and based on the second content identification code and the first content identification code, The first intermediate information corresponding to the content identification codes respectively determines whether the content to be monitored is compliant; wherein the first intermediate information is a summary of the identity encoding information of the played content.
- the first intermediate information of the second content identification code and the first content identification may be compared. Whether the first intermediate information of the code is consistent, if consistent, then the content to be monitored is compliant, if not, then the content to be monitored is non-compliant.
- the first intermediate information is the intermediate information in the process of converting the identity encoding information into the content identification code. It can be understood as a summary of the identity encoding information. For example, the identity encoding information undergoes the first conversion and the second conversion. Even if the content identification code is obtained after the third and more conversions, the first intermediate information may be the intermediate information obtained from the second and third conversions. In this way, by first comparing the first intermediate information, multiple inverse conversions of the second content identification code and the first content identification code can be avoided, thereby improving monitoring efficiency.
- S45 Verification based on the second intermediate information corresponding to the second content identification code and the first content identification code.
- the first content identification code and the second content identification code also include a hidden area, and the hidden area is used to carry the second intermediate information corresponding to the identity encoding information. Then it can be determined that the second content identification code and the first content identification code respectively correspond to whether the second intermediate information is completely consistent.
- both the second intermediate information and the first intermediate information can be intermediate products in the process of converting the identity encoding information into the content identification code, and the second intermediate information can be produced earlier than the first intermediate information.
- the first intermediate information is generated based on the second intermediate information. Therefore, the second intermediate information can carry more original information, or can be understood as the second intermediate information is closer to the original identity encoding information. Therefore, based on The second intermediate information can accurately determine whether the content to be monitored is compliant. in:
- first intermediate information corresponding to the second content identification code and the first content identification code are completely consistent, it is determined that the content to be monitored is compliant, and the process ends.
- the hidden areas of the second content identification code and the first content identification code are respectively analyzed to obtain the second content identification code. and second intermediate information respectively corresponding to the first content identification code; and based on the second intermediate information respectively corresponding to the second content identification code and the first content identification code, it is determined whether the content to be monitored is compliant.
- the two second intermediate information can be compared bit by bit; where, when the two second intermediate information are completely consistent, it is determined that the content to be monitored is compliant; when the two second intermediate information are not completely consistent, Under this condition, secondary verification can be performed, or it can be directly determined to be non-compliant and the process ends.
- Alarm information corresponding to the level is generated to indicate what type of inconsistency is, for example, whether it is an inconsistency in the playback period or an inconsistency in the display device used for playback.
- different fields of the identity encoding information can be obtained by encoding different content attributes of the predetermined content, that is to say, different fields represent different attributes of the predetermined content, such as name attributes, playback period attributes, and playback devices.
- the device identification attributes can be reflected in different fields.
- the inconsistent fields of the two second intermediate information and the content attributes identified by the inconsistent fields can be obtained; based on the content attributes identified by the inconsistent fields, generate An alarm signal corresponding to the level; wherein the content attribute includes at least one of the identification attribute of the display device and the playback period attribute.
- alarm information of corresponding levels can be generated based on the content attributes corresponding to the inconsistent fields.
- different content attributes can correspond to different levels of alarm information.
- Alarm for example, the identification attribute of the display device can be set to a low-level alarm, the playback period attribute can be set to a high-level alarm, in practice, the name of the content can also be set to a medium-level alarm, etc., of course, only The description is illustrative and does not represent a limitation on the present disclosure.
- the alarm signal when the alarm signal is text-based information, the alarm signal can carry content attributes corresponding to inconsistent fields to indicate in which aspect of the display device non-compliant playback occurs.
- the first content identification code can also be directly converted into the original encoded identity encoding information. In this way, it can be determined based on the identity encoding information corresponding to the first content identification code and the second content identification code. Whether the content to be monitored is compliant or not is verified based on the original identity coding information.
- the identity encoding information corresponding to the first content identification code can be directly extracted from the hidden area and the second intermediate information is then processed for thirty seconds. After binary conversion, the original identity encoding information is obtained.
- the second content identification code can be used as a comparison, the compliance of the content being played on the display device can be monitored in a more fine-grained manner from dimensions such as playback period and display device. Therefore, it can be applied to certain In the media resource management system of the advertising business, it is necessary to accurately monitor whether the playback period of the advertising content is correct and whether the display device for playing the advertising content is correct.
- illegal content that has been invaded by a third party and played can be detected in a timely manner, which improves the efficiency of content monitoring and broadens the application scenarios of content monitoring.
- the content identification code is an identification code corresponding to the predetermined content that needs to be displayed on the display device. It is distinguished from the first content identification code corresponding to the content being played on the display device.
- the identification code corresponding to the predetermined content is called the second Content identification code.
- the content identification code may be an image type identification code, and the content identification code may hide the identity encoding information of the currently playing content through color information on the image. That is to say, in one case, the identity encoding information of the text type can be converted into the image type information to uniquely represent the text content through the image.
- This method can be understood as a class conversion of information, that is, the information is transformed from a The format category is converted to another format category, while the meaning of the information remains unchanged, thereby achieving a high level of confidentiality of the information, avoiding information leakage, and thus maintaining information security.
- FIG. 5 a schematic diagram of the principle of converting text-type identity coding information into an image-type content identification code according to the present disclosure is shown. In conjunction with FIG. 5 , the generation process of the second content identification code of the present disclosure is explained. .
- the identity encoding information can be obtained by encoding the multi-dimensional content attributes of the subscription content respectively.
- the multi-dimensional content attributes of the predetermined content can be obtained; the multi-dimensional content attributes are encoded in hexadecimal to obtain the identity coding information of the second predetermined number of digits.
- the multi-dimensional content attributes include one or more of the name of the predetermined content, the playback period, and the device identification of the display device that plays the predetermined content.
- the identity information coding can be a 26-bit hexadecimal coding, and the starting 0-5 bits represent the content.
- the name of The number can also be customized and numbered when making plans. For example, "1" represents one window; “2" represents two windows, distributed horizontally; “3” represents two windows, distributed vertically; “4" represents Three windows, one row, distributed horizontally; "5" represents three windows, two rows, two windows in the first row, one window in the second row; "6” represents three windows, two rows, one window in the first row, There are two windows in the second row; "7” represents three windows, three rows, distributed vertically.
- the 8th digit in the identity code information is the program type of the content, which specifically includes advertising videos, advertising posters, bus route maps, bus timetables, notices and announcement texts, and other information that can be displayed on the media terminal.
- Bits 9-14 indicate the program ID of the content; bits 15-17 indicate the device ID of the display device to be played; bits 18-21 indicate the starting time when the content is to be played; bits 22-25 indicate that the content is to be received The end time of playback.
- Optional example 1 Map the value of each bit in the first intermediate information corresponding to the identity encoding information to the grayscale value of the corresponding check point in the transparent image to obtain the content identification code.
- Optional Example 2 Based on the content identification code obtained in Optional Example 1, add a new hidden area in the transparent image, and map the second intermediate information corresponding to the identity encoding information to the hidden area, thereby obtaining two types of intermediate information. content identification code.
- the identity encoding information can be iteratively encoded in multiple ways to obtain intermediate information under each encoding; the first intermediate information in the intermediate information corresponding to the multiple encodings is mapped to multiple intermediate information on the predetermined transparent image. check points to obtain the second content identification code.
- the first intermediate information is decimal grayscale information, and values at different positions in the first intermediate information correspond to check points at different positions.
- the identity encoding information can be iteratively converted to binary first; then the intermediate information obtained by binary conversion can be converted to hexadecimal; and the intermediate information obtained by hexadecimal conversion can be converted to decimal to obtain decimal. Grayscale information.
- the decimal grayscale information is the first intermediate information.
- the first intermediate information can have a predetermined number of digits. For example, it is 32-bit decimal information. In this way, the first intermediate information has 32 numerical values, and the value of each numerical value is Between 0-9.
- the predetermined transparent image may contain multiple check points, and the number of check points may be the same as the number of first intermediate information, such as 32 check points.
- the value of each bit in the first intermediate information can be used as the gray value of the check point at the corresponding position, thereby assigning a gray scale to the check point on the predetermined transparent image.
- the second content identification code It is an image with multiple check points, and the check points are grayscale and other areas are transparent.
- the first intermediate information is from The values from left to right correspond to the check points from top to bottom and from left to right on the preset transparent image.
- the multiple check points of the preset transparent image can be arranged in an array, in columns, or in a row, or in other shapes.
- the check points may be circular points, rectangular points, or points of other shapes, which are not limited here.
- the grayscale of the check points in the first content identification code and the second content identification code can be separately The value is parsed, and then, according to the positional correspondence between the check point and the bit, the grayscale values are combined to reversely parse out the first intermediate information, so as above, based on the two first intermediate information, it is judged to be Monitor content for compliance.
- the content identification code can carry an information hiding area
- a hidden area can be set in the predetermined transparent image, and then, using The information hiding area carries the second intermediate information corresponding to the identity encoding information.
- the hidden area can be set below the predetermined transparent image.
- Figure 5 is only an exemplary illustration. In practice, the hidden area can be set above, left or right of the predetermined transparent image.
- the second intermediate information among the intermediate information corresponding to each of the multiple codes can be obtained; the second intermediate information is produced earlier than the first intermediate information, and the second intermediate information is binary information; and based on each of the second intermediate information
- the encoded value of the bit updates the grayscale value of the information hidden area.
- the second intermediate information may be information obtained by binary conversion of the identity encoding information.
- the identity encoding information may be encoded using ASCII, so that each character contains an 8-bit binary encoding. Therefore, The 0 and 1 of each bit in the binary coded information can be used as the gray level of a small interval in the hidden area, thereby obtaining the hidden area.
- the cell intervals may be obtained by dividing the hidden area according to the number of bits of the second intermediate information, and different cell intervals correspond to values at different positions of the second intermediate information.
- each character contains 8-bit binary encoding
- the second intermediate information has 208 bits, and the value of each bit is either 0 or 1. Then the value of 208 bits can be used as the gray value of the corresponding small interval in the hidden area.
- the first content identification code and the second content identification code can be determined respectively. Analyze the grayscale value between each small interval in the hidden area, and then combine the grayscale values according to the positional correspondence between the small interval and the bit, so as to reversely analyze the second intermediate information, so as above, Based on the two second intermediate information, it is determined whether the content to be monitored is compliant.
- the identity encoding information is converted into multiple encodings to obtain the content identification code
- the security of the content identification code can be improved, making the content identification code less likely to be cracked by a third party, thus enhancing the inspection of the media resource management system. Inspection safety protection.
- the identity encoding information is converted into the content identification code of the image category, the information is converted from one format category to another format category, while the meaning of the information remains unchanged, thereby achieving a high level of confidentiality of the information. This further increases the difficulty of cracking the content identification code.
- the verification is first performed based on the first intermediate information, because the first intermediate information only needs to be reversely parsed very few times. , thus improving the monitoring speed; later, when the verification based on the first intermediate information fails, verification will be based on the second intermediate information. Since the second intermediate information is closer to the original identity encoding information, it contains sufficient information, so that The verification based on the second intermediate information can improve the monitoring accuracy. In the first-level verification stage, playback abnormalities of the content to be monitored can be accurately detected, thus ensuring the monitoring accuracy while taking into account the monitoring speed.
- FIG. 7 shows the complete flow diagram of the first-level verification in this example.
- Figure 8. showing the complete flow diagram of the second-level verification in this example.
- the content publishing server generates a corresponding content identification code for each content.
- each character includes 8-bit binary encoding to obtain binary intermediate information (second intermediate information); use MD5 or SHA1 encoding for the binary intermediate information to generate 32-bit hexadecimal Information (first intermediate information), convert hexadecimal information into decimal grayscale information, and use the value of each bit in the grayscale information as the grayscale value of the corresponding check point on the predetermined transparent image; convert the second intermediate information
- first intermediate information 32-bit hexadecimal Information
- convert hexadecimal information into decimal grayscale information convert the value of each bit in the grayscale information as the grayscale value of the corresponding check point on the predetermined transparent image
- convert the second intermediate information Each value of each value is used as a grayscale and placed in the information hidden area.
- the broadcast control host sends the predetermined content 1 and the corresponding content identification code to the LED display screen for display, so that the LED display screen displays the content identification when playing the predetermined content 1. code.
- the broadcast control host obtains a screenshot of the display page from the LED display screen, extracts the first content identification code from the screenshot of the display page, and obtains the second content identification code corresponding to the playback period locally.
- the broadcast control host parses the grayscale value of each check point from the first content identification code, and then combines the grayscale values according to the position corresponding relationship to obtain decimal information, and then converts the decimal information into decimal Hexadecimal information (the first intermediate information corresponding to the first content identification code); similarly, the hexadecimal information (the first intermediate information corresponding to the second content identification code corresponding to the second content identification code) stored during the conversion process can be directly obtained. an intermediate information).
- S75 Compare the two hexadecimal information to see if they are consistent. If they are consistent, end the process and output the success code 1. Upload a screenshot of the display page of the content publishing server to keep the bottom; if they are inconsistent, continue to start from the first content identification code.
- the binary intermediate information (second intermediate information) is parsed out from the hidden area, and the binary intermediate information is converted back into hexadecimal identity encoding information.
- S76 Compare whether the two identity coding information are consistent. If they are completely consistent, the process ends and the success code 2 is output. Upload a screenshot of the display page of the content publishing server to keep the bottom; if they are inconsistent, compare the two identity codes bit by bit. information.
- S77 Compare whether the fields corresponding to the device identification of the display device are consistent, that is, whether the content attributes of the device identification are consistent. If they are consistent, the alarm code 1 will be output, and a screenshot of the display page will be uploaded to the content publishing server for secondary verification. ; If they are inconsistent, continue to compare whether the fields corresponding to the playback period are consistent, that is, whether the content attributes of the playback period are consistent. If they are consistent, alarm code 2 will be output, and a screenshot of the display page will be uploaded to the content publishing server for secondary verification. ; If they are inconsistent, alarm code 3 will be output, and a screenshot of the display page will be uploaded to the content publishing server for secondary verification.
- the level of alarm code 1 is lower than the level of alarm code 2
- the level of alarm code 2 is lower than the level of alarm code 3.
- the broadcast control host sends the screenshot of the display page and the corresponding alarm code to the content publishing server;
- S83 When the content publishing server detects warning code 3, it first detects screen anomalies such as black screen and blurry screen with incomplete playback. For example, input the screenshot of the display page into the first detection model. If the result output by the first detection model If the screen is black/distorted/incomplete playback, etc., which is not in line with the expected situation, report the information to the administrator for maintenance of the LED display.
- screen anomalies such as black screen and blurry screen with incomplete playback. For example, input the screenshot of the display page into the first detection model. If the result output by the first detection model If the screen is black/distorted/incomplete playback, etc., which is not in line with the expected situation, report the information to the administrator for maintenance of the LED display.
- S84 After eliminating screen abnormalities such as black screens and blurred screens, detect pornography, explosions, and sensitive information. For example, input screenshots of the display page into the second detection model.
- the second detection model can use computer vision algorithms and OCR and other technologies to extract keywords. , if the result output by the second detection model has the above-mentioned illegal content, the broadcast control host is instructed to directly turn off the LED display and alert the administrator.
- the playback content monitoring method of the present disclosure is used to ensure the security of the playback content distributed by the media resource management system to a variety of remote screens (LCD display screens, LED displays, and display screens on other types of display terminals).
- the broadcast control host deployed at the edge first performs a first-level verification. If the first-level verification passes, the corresponding processing will be carried out directly, and the processing results will be kept on the content server. If the first-level verification fails, a screenshot of the displayed page will be fed back to the content server for more intelligent second-level verification to make corresponding decisions.
- Using edge computing and hierarchical verification strategies can improve the utilization rate of edge devices (broadcast control hosts). It helps the monitoring system respond quickly, while reducing the impact of network concurrency and packet loss, and reducing the computing pressure on the content publishing server.
- encryption algorithm SDK and verification algorithm SDK disclosed in this disclosure can be deployed in the processor or memory of any device.
- Media resource management system distribution can also be sent to mobile phones, tablets, laptops and other application terminals.
- compliance testing can also be performed on the playback content distributed by the media resource management system only to terminals such as mobile phones, tablets, and laptops.
- the present disclosure also provides a playback control host, which is used to execute the method for monitoring playback content.
- the broadcast control host can be deployed at the edge of the media resource management system and directly connected to various types of display devices.
- One broadcast control host can connect to one or more display devices.
- the present disclosure also provides a broadcast monitoring system.
- a schematic system structure diagram of a broadcast monitoring system is shown. As shown in Figure 9, it includes: a content publishing server, and a broadcast monitoring system. At least one broadcast control host connected to the content publishing server, and a display device connected to each of the broadcast control hosts;
- the content publishing server is used to send the content to be played to the corresponding broadcast control host;
- the playback control host is used to send the content to be played to the display device, and to execute the method for monitoring the playback content
- the display device is used to play the content to be played.
- the content publishing server can be the backend server of the content publishing platform, and the content publishing platform can be a web application or a mobile application, such as an APP.
- the broadcast control host can perform the first-level verification and the second-level verification described in the above embodiment. In yet another optional example, the broadcast control host can perform the first-level verification and second-level verification described in the above embodiment. For the first-level verification, the content publishing server can perform the second-level verification in the above embodiment.
- FIG. 9 shows the situation of three broadcast control hosts and five display devices.
- the number of broadcast control hosts and the number of display devices may not be limited by FIG. 9 .
- the present disclosure also provides a monitoring device for playing content.
- a schematic structural diagram of a monitoring device according to an embodiment of the present disclosure is shown. As shown in Figure 10, the device includes:
- the screenshot acquisition module 1001 is used to obtain a screenshot of the display page from the display device during the current monitoring period.
- the screenshot of the display page is obtained by taking a screenshot of the content to be monitored currently played by the display device;
- the identification code acquisition module 1002 is used to extract a first content identification code from the display page screenshot, and the first content identification code is used to identify the played content;
- the verification module 1003 is configured to determine whether the content to be monitored is compliant based on the first content identification code.
- the verification module 1003 includes:
- a first verification unit configured to perform a first-level verification based on the first content identification code, and the first-level verification is used to determine whether there is an abnormality in the content to be monitored;
- the second verification unit is used to perform a second-level verification on the screenshot of the display page if the first-level verification fails, and the second-level verification is used to determine the abnormal category of the content to be monitored;
- a judgment unit configured to determine whether the content to be monitored is compliant based on the results of the secondary verification.
- the second verification unit is specifically configured to perform picture abnormality detection on the screenshot of the display page, and determine whether there is a first type of abnormal picture and/or a second type of abnormal picture in the content to be monitored;
- the first type is the type of screen exception
- the second type is the type containing content exception;
- the judgment unit is specifically configured to determine that the content to be monitored is compliant if there is an abnormal picture of the first type; and to determine that the content to be monitored is not compliant if there is an abnormal picture of the second type.
- the second verification unit includes:
- the first detection subunit is used to input the screenshot of the display page into the first detection model, and determine whether the first type of abnormal screen exists in the content to be monitored according to the result output by the first detection model;
- the second detection subunit is used to input the screenshot of the display page into the second detection model, and determine whether there is a second type of abnormal screen in the content to be monitored according to the result output by the second detection model; wherein,
- the first detection model is obtained by training the first neural network using multiple images carrying screen abnormalities as training samples, and the second detection model is obtained by training the second neural network using multiple images carrying abnormal content as training samples.
- the network is trained.
- the second verification unit includes:
- a sending subunit configured to send the screenshot of the display page to the content publishing server, so that the content publishing server can perform the secondary verification
- the determination unit is specifically configured to determine whether the content to be monitored is compliant based on the secondary verification result returned by the content publishing server.
- the device also includes the following modules:
- a shutdown module configured to shut down the display device to prohibit the display device from playing content when it is determined that the content to be monitored is not compliant
- an alarm module configured to send an alarm signal to indicate that the content to be monitored is not compliant when it is determined that the content to be monitored is not compliant.
- the device also includes:
- a coding information generation module configured to generate identity coding information corresponding to the scheduled content to be sent to the display device for playback
- a content identification code generation module configured to generate a second content identification code of the predetermined content based on the identity encoding information
- a sending module configured to send the predetermined content and the second content identification code to the display device, so that the display device displays the second content identification code on the display page when playing the predetermined content. ; Wherein, the second content identification code is used as a comparison with the first content identification code to determine whether the content to be monitored is compliant.
- the verification module 1003 includes:
- a second content identification code acquisition unit is configured to obtain a second content identification code corresponding to the playback period to be monitored, wherein the second content identification code is the same as the scheduled content to be played during the playback period to be monitored. correspond;
- a verification unit configured to determine whether the content to be monitored is compliant based on the first content identification code and the second content identification code.
- the verification module 1003 includes:
- a parsing unit configured to parse the second content identification code and the first content identification code respectively to obtain the first intermediate information corresponding to the second content identification code and the first content identification code respectively; wherein , the first intermediate information is the intermediate information in the process of converting the identity coding information corresponding to the content into the content identification code;
- a first judgment unit configured to determine whether the content to be monitored is compliant based on the first intermediate information respectively corresponding to the second content identification code and the first content identification code;
- the first intermediate information is a summary of the identity encoding information of the played content.
- the second content identification code also includes a hidden area, the hidden area is used to carry second intermediate information corresponding to the identity encoding information, and the first judgment unit is specifically used to:
- Information wherein the second intermediate information is produced before the first intermediate information;
- the step of determining whether the content to be monitored is compliant based on the second intermediate information corresponding to the second content identification code and the first content identification code includes:
- the device also includes:
- a content attribute acquisition module configured to acquire the inconsistent fields of the two second intermediate information and the content attributes identified by the inconsistent fields when the two second intermediate information are not completely consistent
- An alarm signal generation module is configured to generate an alarm signal of a corresponding level based on the content attributes identified by the inconsistent fields; wherein the content attributes include at least one of an identification attribute of the display device and a playback period attribute.
- the content identification code generation module includes:
- a coding unit used to iteratively perform multiple codings on the identity coding information to obtain intermediate information under each coding
- a first mapping unit configured to map the first intermediate information in the intermediate information corresponding to the plurality of codes to a plurality of check points on the predetermined transparent image to obtain the second content identification code
- the first intermediate information is decimal grayscale information, and grayscale values at different positions in the first intermediate information correspond to check points at different positions.
- the coding unit includes;
- the first conversion subunit is used to perform binary conversion of the identity encoding information
- the second conversion subunit is used to convert the intermediate information obtained by binary conversion into hexadecimal
- the third conversion subunit is used to convert the intermediate information obtained by hexadecimal conversion into decimal to obtain decimal grayscale information.
- an information hiding area is set on the predetermined transparent image, and the information hiding area is used to carry the second intermediate information corresponding to the identity encoding information.
- the device further includes:
- the second intermediate information acquisition module is used to acquire the second intermediate information among the intermediate information corresponding to each of the plurality of codes; the second intermediate information is produced earlier than the first intermediate information, and the second intermediate information is binary information;
- a mapping module configured to update the grayscale value of the information hidden area based on the encoding value of each bit in the second intermediate information.
- the encoding information generation module includes:
- a content attribute acquisition unit configured to acquire multi-dimensional content attributes of the predetermined content.
- the multi-dimensional content attributes include the name of the predetermined content, the playback period, and the device identification of the display device that plays the predetermined content. one or more;
- An encoding unit configured to perform hexadecimal encoding on the multi-dimensional content attributes to obtain a second predetermined number of digits of identity encoding information.
- the present disclosure also provides a computer-readable storage medium, the computer program stored therein causes the processor to execute the method for monitoring playback content.
- the present disclosure also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor.
- the processor executes, the described playback of content is realized. Monitoring methods.
- any reference signs placed between parentheses shall not be construed as limiting the claim.
- the word “comprising” does not exclude the presence of elements or steps not listed in a claim.
- the word “a” or “an” preceding an element does not exclude the presence of a plurality of such elements.
- the present disclosure may be implemented by means of hardware comprising several different elements and by means of a suitably programmed computer. In the element claim enumerating several means, several of these means may be embodied by the same item of hardware.
- the use of the words first, second, third, etc. does not indicate any order. These words can be interpreted as names.
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Strategic Management (AREA)
- Finance (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Databases & Information Systems (AREA)
- Game Theory and Decision Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Economics (AREA)
- Marketing (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Two-Way Televisions, Distribution Of Moving Picture Or The Like (AREA)
Abstract
一种播放内容的监测方法、主机、***、装置、介质及设备,属于通信技术领域,该方法包括:在当前的监测时段,从显示设备获取显示页面截图,显示页面截图是对显示设备的显示页面进行截图得到的;从显示页面截图中提取第一内容标识码,第一内容标识码用于标识当前所播放的待监测内容;基于第一内容标识码,确定待监测内容是否合规,上述方法可以由播控主机执行,从而实现在边缘端对显示屏的在播内容进行监测,同时采用分级校验的手段,在基于第一内容标识码,确定待监测内容是否合规时,由播控主机确定待监测内容是否异常,若异常可以再由内容发布服务器基于显示页面截图判定待监测内容是否合规。
Description
本公开涉及通信技术领域,特别是涉及一种播放内容的监测方法、主机、***、装置、介质及设备。
随着物联网技术的兴起,媒体资源管理***应运而生,其中,媒体资源管理***广泛应用在交通、金融、商场、超市等场景中,媒体资源管理***通常包括显示设备,该显示设备用于播放指定的媒体内容,如广告、影视剧、促销信息等。
在实际的使用场景里,需要对显示设备所显示的内容进行监控,相关技术中,运营管理人员需要人工巡查监视显示设备上的播放状态、播放内容是否合规合规。但是此种方式存在成本高、效率低的问题。
概述
本公开提供了一种播放内容的监测方法,所述方法包括:
在当前的监测时段,从显示设备获取显示页面截图,所述显示页面截图是对所述显示设备的显示页面进行截图得到的;
从所述显示页面截图中提取第一内容标识码,所述第一内容标识码用于标识当前所播放的待监测内容;
基于所述第一内容标识码,确定所述待监测内容是否合规。
在一种可选的示例中,基于所述第一内容标识码,确定所述待监测内容是否合规,包括:
基于所述第一内容标识码,进行一级校验,所述一级校验用于确定所述待监测内容是否存在异常;
在所述一级校验不通过的情况下,对所述显示页面截图进行二级校验,所述二级校验用于确定所述待监测内容的异常类别;
基于所述二级校验的结果,确定所述待监测内容是否合规。
在一种可选的示例中,对所述显示页面截图进行二级校验,包括:
对所述显示页面截图进行画面异常检测,确定所述待监测内容是否存在第一类型的异常画面和/或第二类型的异常画面;第一类型为屏幕异常的类型,第二类型为包含内容异常的类型;
基于所述二级校验的结果,确定所述待监测内容是否合规,包括:
若存在所述第一类型的异常画面,则确定所述待监测内容合规;
若存在第二类型的异常画面,则确定所述待监测内容不合规。
在一种可选的示例中,对所述显示页面截图进行画面异常检测,确定所述待监测内容是否存在第一类型的异常画面和/或第二类型的异常画面,包括:
将所述显示页面截图输入至第一检测模型,并根据所述第一检测模型输出的结果确定所述待监测内容是否存在所述第一类型的异常画面;
和/或,将所述显示页面截图输入至第二检测模型,并根据所述第二检测模型输出的结果确定所述待监测内容是否存在第二类型的异常画面;
其中,所述第一检测模型是以多个携带有屏幕异常的图像为训练样本对第一神经网络进行训练得到的,所述第二检测模型是以多个携带异常内容的图像为训练样本对第二神经网络进行训练得到的。
在一种可选的示例中,在所述一级校验不通过的情况下,对所述显示页面截图进行二级校验,包括:
将所述显示页面截图发送给内容发布服务器,以使所述内容发布服务器进行所述二级校验;
基于二级校验的结果,确定所述待监测内容是否合规,包括:
基于所述内容发布服务器返回的二级校验的结果,确定所述待监测内容是否合规。
在一种可选的示例中,在确定所述待监测内容不合规的情况下,所述方法还包括以下至少之一:
关闭所述显示设备,以禁止所述显示设备进行内容播放;和/或
发送报警信号,以指示所述待监测内容不合规。
在一种可选的示例中,在所述在当前的监测时段,从显示设备获取显示页面截图的步骤之前,所述方法还包括:
对待发送给所述显示设备进行播放的预定内容,生成与所述预定内容对 应的身份编码信息;
基于所述身份编码信息,生成所述预定内容对应的第二内容标识码;
将所述预定内容与所述第二内容标识码发送给所述显示设备,以使所述显示设备在播放所述预定内容时,在显示页面展示所述第二内容标识码;
其中,所述第二内容标识码作为所述第一内容标识码的对照,以用于确定所述待监测内容是否合规。
在一种可选的示例中,基于所述第一内容标识码,确定所述待监测内容是否合规,包括:
获取与所述待监测的播放时段对应的第二内容标识码,其中,所述第二内容标识码与在所述待监测的播放时段进行播放的预定内容对应;
基于所述第一内容标识码和所述第二内容标识码,确定所述待监测内容是否合规。
在一种可选的示例中,基于所述第一内容标识码,确定所述待监测内容是否合规,包括:
分别对所述第二内容标识码和所述第一内容标识码进行解析,得到所述第二内容标识码和所述第一内容标识码分别对应的第一中间信息;其中,所述第一中间信息是将内容对应的身份编码信息转换为内容标识码的过程中的中间信息;
基于所述第二内容标识码和所述第一内容标识码分别对应的第一中间信息,确定所述待监测内容是否合规;
其中,所述第一中间信息为所播放的内容的身份编码信息的摘要。
在一种可选的示例中,所述第二内容标识码还包括隐藏区,所述隐藏区用于携带身份编码信息对应的第二中间信息,所述基于所述第二内容标识码和所述第一内容标识码分别对应的第一中间信息,确定待监测内容是否合规的步骤,包括:
确定所述第二内容标识码和所述第一内容标识码分别对应的第一中间信息是否完全一致;
若否,则分别对所述第二内容标识码和所述第一内容标识码的隐藏区进行解析,得到对所述第二内容标识码和所述第一内容标识码各自对应的第二中间信息;其中,所述第二中间信息在所述第一中间信息之前产出;
基于所述第二内容标识码和所述第一内容标识码各自对应的第二中间信息,确定所述待监测内容是否合规。
在一种可选的示例中,基于所述第二内容标识码和所述第一内容标识码各自对应的第二中间信息,确定所述待监测内容是否合规,包括:
按位比对两个所述第二中间信息;
在两个所述第二中间信息完全一致的情况下,确定所述待监测内容合规;
在两个所述第二中间信息不完全一致的情况下,确定所述待监测内容不合规。
在一种可选的示例中,在两个所述第二中间信息不完全一致的情况下,所述方法还包括:
获取两个所述第二中间信息不一致的字段,以及所述不一致的字段所标识的内容属性;
基于所述不一致的字段所标识的内容属性,生成对应等级的报警信号;其中,所述内容属性包括显示设备的标识属性、播放时段属性中的至少一者。
在一种可选的示例中,基于所述身份编码信息,生成所述预定内容的第二内容标识码,包括:
对所述身份编码信息迭代进行多种编码,得到每种编码下的中间信息;
将所述多种编码各自对应的中间信息中的第一中间信息映射为预定透明图像上的多个校验点,得到所述第二内容标识码;
其中,所述第一中间信息为十进制的灰度信息,所述第一中间信息中不同位置的数值对应不同位置的校验点。
在一种可选的示例中,所述对所述身份编码信息迭代进行多种编码,得到每种编码下的中间信息的步骤,包括;
将所述身份编码信息进行二进制转换;
将所述二进制转换得到的中间信息进行十六进制转换;
将十六进制转换得到的中间信息进行十进制转换,得到十进制的灰度信息。
在一种可选的示例中,所述预定透明图像上设置信息隐藏区,所述信息隐藏区用于携带所述身份编码信息对应的第二中间信息,所述方法还包括:
获取所述多种编码各自对应的中间信息中的第二中间信息;所述第二中 间信息早于所述第一中间信息产出,所述第二中间信息为二进制的信息;
基于所述第二中间信息中每位的编码值,更新所述信息隐藏区的灰度值。
在一种可选的示例中,所述生成与所述预定内容对应的身份编码信息的步骤,包括:
获取所述预定内容的多维度的内容属性,所述多维度的内容属性包括所述预定内容的名称、播放时段以及播放所述预定内容的显示设备的设备标识中的一者或多者;
分别对所述多维度的内容属性进行三十二进制编码,得到所述身份编码信息。
本公开还提供一种播控主机,所述播控主机用于执行所述的播放内容的监测方法。
本公开还提供一种监播***,包括:内容发布服务器,与所述内容发布服务器连接的至少一个播控主机,以及与每个所述播控主机连接的显示设备;
所述内容发布服务器,用于将待播放的内容发送给对应的播控主机;
所述播控主机,用于将所述待播放的内容发送给所述显示设备,并用于执行所述的播放内容的监测方法;
所述显示设备,用于播放所述待播放的内容。
基于相同的发明构思,本公开还提供了一种计算机可读存储介质,其存储的计算机程序使得处理器执行所述的播放内容的监测方法。
基于相同的发明构思,本公开还提供了一种电子设备,包括存储器、处理器及存储在存储器上并可在处理器上运行的计算机程序,所述处理器执行时实现所述的播放内容的监测方法。
本公开还提供一种播放内容的监测装置,所述装置包括:
截图获取模块,用于在当前的监测时段,从显示设备获取显示页面截图,所述显示页面截图是对所述显示设备当前播放的待监测内容进行截图得到的;
标识码获取模块,用于从所述显示页面截图中提取第一内容标识码,所述第一内容标识码用于标识所播放的内容;
验证模块,用于基于所述第一内容标识码,确定所述待监测内容是否合规。
本公开实施例的技术方案,可以在当前的监测时段,从显示设备获取显 示页面截图,并从显示页面截图中提取第一内容标识码,之后基于第一内容标识码,确定待监测内容是否合规。
由于显示页面截图是对显示设备的显示页面进行截图得到的,而第一内容标识码用于标识当前所播放的待监测内容,这样便可以基于内容标识码确定显示设备当前播放的内容是何种内容,从而可以基于该第一内容标识码确定待监测内容是否合规。一方面,可以只需对显示设备的显示页面进行截图,即可基于从显示页面截图中提取的内容标识码进行播放内容的合规性验证,从而无需人工巡检,降低了巡检成本,提高了巡检效率。另一方面,通过第一内容标识码可以确定待监测内容是否为指定播放的内容,以及可以确定待监测内容的播放页面是否出现异常,由此,可以提高对播放内容巡检的准确性。
上述说明仅是本公开技术方案的概述,为了能够更清楚了解本公开的技术手段,而可依照说明书的内容予以实施,并且为了让本公开的上述和其它目的、特征和优点能够更明显易懂,以下特举本公开的具体实施方式。
附图简述
为了更清楚地说明本公开实施例或相关技术中的技术方案,下面将对实施例或相关技术描述中所需要使用的附图作一简单地介绍,显而易见地,下面描述中的附图是本公开的一些实施例,对于本领域普通技术人员来讲,在不付出创造性劳动的前提下,还可以根据这些附图获得其他的附图。需要说明的是,附图中的比例仅作为示意并不代表实际比例。
图1示意性地示出了本公开的一种播放内容的监测方法所应用的通信环境示意图;
图2示意性地示出了本公开的一种播放内容的监测方法的步骤流程示意图;
图3示意性地示出了本公开的又一种播放内容的监测方法的流程示意图;
图4示意性地示出了本公开的一种播放内容的监测方法的整体流程示意图;
图5示意性地示出了本公开的将文本类型的身份编码信息转换为图像类型的内容标识码的原理示意图;
图6示意性示出了本公开的一种示例中身份编码信息的编码示意图;
图7示意性示出了本公开的一级校验的完整流程示意图;
图8示意性示出了本公开的二级校验的完整流程示意图;
图9示意性示出了本公开的一种监播***的框架结构示意图;
图10示意性示出了本公开的一种播放内容的监测装置的结构示意图。
详细描述
为使本公开实施例的目的、技术方案和优点更加清楚,下面将结合本公开实施例中的附图,对本公开实施例中的技术方案进行清楚、完整地描述,显然,所描述的实施例是本公开一部分实施例,而不是全部的实施例。基于本公开中的实施例,本领域普通技术人员在没有作出创造性劳动前提下所获得的所有其他实施例,都属于本公开保护的范围。
媒体资源管理(Media Asset Management,MAM)***是建立在多媒体、网络、数据库和数字存储等先进技术基础上的一个对各种媒体及内容(如视/音频资料、文本文件、图表等)进行数字化存储、管理以及应用的总体解决方案,包括数字媒体的采集、编目、管理、传输和编码转换等所有环节。其主要是满足媒体资源拥有者收集、保存、查找、编辑、发布各种信息的要求,为媒体资源的使用者提供访问内容的便捷方法,实现对媒体资源的高效管理,大幅度提高媒体资源的价值。
相关技术中,媒体资源管理***的业务痛点为:内容发布后,需要监控显示设备端是否按照预定计划在播,这是因为显示设备可能被第三方入侵,使得显示设备被第三方播放涉黄涉毒或者政治敏感内容等非法内容;当然,在一些情况下,还需要及时监测显示设备是否黑屏、花屏,在播内容是否应该在当前显示设备播放,以及在播内容的播放时间段是否正确等。
然而,当前普遍通过人工巡点方式进行监播,成本高,效率低,遇到突发状况处置不及时。
在一种处理方案中,可以为每个显示设备配套监播相机,将监播相机对准显示设备的显示屏幕进行录像,从而基于录制的图像通过AI等技术实现对屏幕内容的自动分析,以此来对显示设备的在播内容进行监测。但此类方式存在如下问题:
第一,需要为每个显示设备配套监播相机,不仅增加了成本,在架设监播相机时,还需要遵循特定的高度和角度等标准来安装相机,且安装相机相关的结构、布线等工程在较多客户现场受到限制,在工程实施上难以落地。
第二,由于对显示屏幕的屏幕内容的自动分析一般需要耗费较多的计算资源,而对于一套媒体资源管理***而言,需要监测的显示设备一般多达几十上百个,显然这样的方式无法满足及时高效且低投入的监测需求。
有鉴于此,本公开为实现对媒体资源管理***中多个显示设备的高效监测,且减小监测成本,提出了一种播放内容的监测方法,该方法的核心构思在于:
为每个待播放的内容生成对应的一个内容标识码,该内容标识码一般唯一表征一个播放的内容,且可以被显示在显示页面随着内容一起播放,这样在进行播放内容的监测时,可以对显示设备的播放页面进行截图,进而从截图中提取该内容标识码,从而基于该内容标识码可以确定显示设备在播的内容是否合规。
参照图1所示,示出了本公开的一种媒体资源管理***的通信环境示意图,如图1所示,包括:内容发布平台、内容发布服务器、多个播控主机以及与每个播控主机连接的显示设备,其中,显示设备可以包括LED显示屏,也可以包括LCD显示屏,当然,实际中也可以包括其他类型的显示设备,如手机、电脑、平板等终端设备的显示屏。
其中,内容发布服务器连接多个播控主机,一个播控主机可以连接一个或多个相同类型的显示设备,或连接多个不同类型的显示设备。
其中,播控主机与显示设备之间可以通过串口/HDMI作为数据的下行通路,以发送待播放的内容,通过网口/USB作为数据的上行通路,以接收显示设备上传的显示页面截图。
在实际运行中,运营人员可以登录内容发布平台发布要播放的内容,指定该待播放的内容对应的显示设备,之后点击发布,则内容发布平台可以将该待播放的内容,以及待播放的内容对应的显示设备的标识一并通过内容发布服务器发送给播控主机,之后由播控主机将该待播放的内容发送给对应的显示设备。
当然,在一些情况下,用户登录内容发布平台进行待播放的内容的发布 时,还可以指定该待播放的内容的播放时段,此种情况下,内容发布服务器可以一并将播放时段发送给播控主机,由播控主机指示该显示设备在对应的播放时段对该待播放的内容进行播放。
其中,本公开的内容可以是指多媒体数据,例如广告、电视节目等。
基于图1所示的通信环境,对本公开的一种播放内容的监测方法进行介绍,参照图2所示,示出了本公开的一种播放内容的监测方法的步骤流程图,其中,本实施例的播放内容的监测方法可以由图1所示的播控主机执行,当然,在其他一些可行的实施例中,也可以由内容发布服务器执行,或者可以由该媒体资源管理***中的其他终端设备执行,该其他终端设备只要与显示设备进行通信连接即可,本公开不对播放内容的监测方法的执行机构进行特别限定。
如图2所示,主要包括以下步骤:
步骤S201:在当前的监测时段,从显示设备获取显示页面截图,所述显示页面截图是对所述显示设备的显示页面进行截图得到的。
本实施例中,当前的监测时段可以是指显示设备进行内容播放的时段,实际中可以是指显示设备的工作时段,如显示设备在一天的10:00-18:00需要播放内容,则当前的监测时段可以是10:00-18:00的时段。此种情况下,可以每间隔预设的时长,便从显示设备获取显示页面截图,以进行适时的内容监测。
或者,在又一种情况中,可以是指在显示设备的工作时段中对每个内容进行播放的时段,如显示设备在一天的10:00-18:00需要播放内容,其中,内容1在10:00-12:00播放,内容2在12:10-13:00播放,则当前的监测时段可以是指当前播放的内容1的时段或当前播放的内容2的时段,此种情况下,可以在每个内容对应的播放时段,都获取显示页面截图,以对每个内容的播放进行监测。
其中,可以在当前的监测时段,从显示设备获取显示页面截图,具体实施时,可以在当前的监测时段的任意时刻,向显示设备发送截图指令,显示设备响应于该截图指令可以对显示设备的显示页面进行截图,从而得到显示页面截图,在显示设备对显示页面进行截图得到显示页面截图时,可以将显示页面截图返回给发送截图指令的播控主机或服务器。
其中,截图的显示页面可以是显示设备的显示屏屏幕,从而保留显示屏屏幕上当前显示的所有内容。在一些可选的示例中,截图的显示页面也可以是显示屏屏幕中当前播放有内容的播放窗口。实际中,可以依据实际情况对全屏截图或者对播放窗口进行截图。
步骤S202:从所述显示页面截图中提取第一内容标识码,所述第一内容标识码用于标识显示设备当前所播放的待监测内容。
本实施例中,需要在显示设备上播放的内容都可以预先生成对应的一个内容标识码,该内容标识码可以用于标识所播放的内容的内容标识、名称、类别、播放时段等属性。实际中,一个内容的内容标识码可以是全局唯一的标识码,以唯一表征一个内容,其中内容标识码可以是文本类型的标识码、字符串类型的标识码或者可以是与二维码类似的标识码。
实际中,在显示设备上播放内容时,一并可以将内容标识码显示在显示设备的显示页面上,这样在对显示页面进行截图时,便可以在截取到所播放的内容的画面时,一并截取到内容标识码。从而在得到显示页面截图时,可以从显示页面截图中提取出第一内容标识码。
具体而言,显示设备在播放内容时,可以将该内容对应的内容标识码显示到显示页面的固定位置,该固定位置依据实际情况可以是显示屏幕的端点位置,也可以是播放窗口中的固定位置,从而无论是对显示页面截图还是对内容的播放窗口截图,均可以保证在显示页面截图中具有第一内容标识码。
其中,在从显示页面截图中提取第一内容标识码时,可以从显示页面截图中与固定位置对应的图像区域中提取出第一内容标识码的图像,并可以对该第一内容标识码的图像进行解析,得到第一内容标识码。
具体而言,若第一内容标识码是文本类型或字符串类型的标识码,可以利用OCR(optical character recognition,光学文字识别)技术进行识别,得到第一内容标识码。若是二维码类型的标识码,可以对该第一内容标识码的图像进行识别,得到其所表征的第一内容标识码。
步骤S203:基于所述第一内容标识码,确定所述待监测内容是否合规。
本实施例中,可以基于第一内容标识码,确定该第一内容标识码所标识的内容,该所述标识的内容可以是显示设备当前播放的待监测内容,从而可以基于该第一内容标识码,确定该待监测内容是否合规。
在一种情况中,若识别出的第一内容标识码残缺,则表征显示页面的显示出现了异常,或者可以表示显示设备的显示页面可能被第三方入侵,由此,可以基于第一内容标识码是否具有残缺,确定待监测内容是否合规;若是有残缺的,则不合规,若是完整的,再进行进一步的监测。
在又一种情况中,若识别出的第一内容标识码完整,则可以检测第一内容标识码是否乱码,乱码可以是指第一内容标识码包括了不应该出现的信息,实际中,当显示设备被第三方入侵时,可能导致这一情况的出现,由此,只要识别到第一内容标识码乱码,则可以确定待监测内容不合规。
本公开中,合规可以是指待监测内容符合平台的内部规则,具体规则可以由平台需求设定,如设定为待监测内容的播放时段符合预定播放时段,待监测内容为指定播放内容,或者待监测内容不包括禁止播放的画面等。
在又一种情况中,在第一内容标识码完整且没有乱码的情况下,可以确定待监测内容合规。
当然,在更进一步的示例中,也可以继续从第一内容标识码解析出待监测内容的内容属性,该内容属性可以包括上述实施例所述的内容的名称、内容标识、播放时段和显示设备的标识等。如此,可以基于解析出的内容属性,确定待监测内容是否合规。
该情况中,若基于内容属性,确定该待监测内容并非在该待监测的播放时段应该播放的内容,或者该待监测内容对应的显示设备的标识,并非是当前的显示设备,则可以确定待监测内容是不合规的。反之,若该待监测内容是在该待监测的播放时段应该播放的内容,或者该待监测内容对应的显示设备的标识,是当前的显示设备,则可以确定待监测内容是合规的。如此,可以使得媒体资源管理***可以进一步巡检显示设备上在播内容的正确性,保证显示设备在一个播放时段播放该播放时段内应该播放的指定内容。
采用本实施例的技术方案,一方面,可以只需对显示设备的当前显示页面进行截图,即可基于从显示页面截图中提取的内容标识码进行播放内容的合规性验证,从而无需人工巡检,降低了巡检成本,提高了巡检效率。另一方面,通过第一内容标识码可以确定待监测内容是否为指定播放的内容,以及可以确定待监测内容的播放页面是否出现异常,由此,可以提高对播放内容巡检的准确性。
再一方面,由于是基于显示页面截图中的第一内容标识码,确定待监测内容是否合规,其中,第一内容标识码一般数据量小,解析分析的计算资源耗费较小,从而实现了以小数据量的信息,实现对内容的快速准确监测。避免了对显示页面截图中与待监测内容有关的画面进行解析分析时,计算资源耗费较多、分析耗时长、准确性严重依赖算法精度(如机器学习中,依赖样本数量多寡和神经网络设计)的问题。
在一种可选的示例中,在确定待监测内容不合规的情况下,还可以关闭显示设备,以禁止显示设备进行内容播放;或者,发送报警信号,以指示待监测内容不合规,或者同时关闭显示设备以及发送报警信号。其中,如上所述,所述合规,也可以是符合平台的内部规则,具体规则可以由平台需求设定。
在该可选的示例中,若依据第一内容标识码确定待监测内容不合规,可以强制关闭显示设备,例如,可以向显示设备发送下电信号,以切断显示设备的电源,从而强制显示设备进行下电,使得显示设备处于关闭状态,无法进行非法的待监测内容的播放。实际中,也可以向上位机和对应的用户发送报警信号,在媒体资源管理***中,上位机可以是图1所述的内容发布平台,对应的用户可以是对显示设备进行巡检的巡检用户。
其中,报警信号可以是报警信息,其可以携带显示页面截图以及显示设备的设备标识,从而指示是哪一显示设备在播放不合规的内容,以及不合规的内容的画面,从而帮助用户基于报警信息对待监测内容进行更进一步的处理。
下面,结合图1所述的通信环境,对本公开的播放内容的监测手段进行介绍,具体来说,本公开的播放内容的监测手段可以包括与监测机制相关的技术手段,以及与信息处理有关的技术手段。其中,监测机制主要体现在对待监测内容确定是否合规的监测流程上,而信息处理主要体现在如何生成内容标识码,以及如何基于内容标识码进行监测的处理手段上。
下面,分别对监测机制和内容处理进行介绍。
1.1,确定待监测内容是否合规的监测机制
实际中,由于本公开需要监测显示设备是否被第三方播放涉黄涉毒或者其他一些敏感内容等非法内容;以及,还需要及时监测显示设备是否黑屏、 花屏,在播内容是否应该在当前显示设备播放,以及在播内容的播放时间段是否正确等,因此将监测任务划分为对待监测内容的异常检测阶段,以及在待监测内容监测出异常后的合规性判定阶段。
其中,异常检测阶段可以是基于第一内容标识码的检测,目的是判断出内容是否出现显示异常以及在播内容是否是指定的内容,合规性判定阶段可以是基于显示页面截图对异常检测阶段所检出的异常,排查出现该异常的原因,具体来说,可以排查导致异常的原因是屏幕显示故障还是第三方入侵。
在一种可选的示例中,参照图3所示,示出了本公开的又一种播放内容的监测方法的流程示意图,如图3所示,具体可以包括以下流程:
S31:在当前的监测时段,从显示设备获取显示页面截图,并从显示页面截图中提取第一内容标识码。其中,第一内容标识码用于标识当前所播放的待监测内容;所述显示页面截图是对所述显示设备的当前屏幕进行截图得到的。
本可选示例中,可以在每个内容的播放时段,从显示设备获取显示页面截图,以在不同的内容所对应的播放时段,对不同的内容进行分别监测,从而可以监测在内容的播放时段,显示设备是否正确播放内容。
具体实施时,可以预先存储每个内容的起始播放时刻和结束播放时刻,从而在每个内容的起始播放时刻与结束播放时刻之间的播放时段,从显示设备获取显示页面截图。
其中,显示页面截图可以是对显示设备的当前屏幕进行截图得到的,其包括了待监测内容的画面,也包括了第一内容标识码。
S32:基于所述第一内容标识码,进行一级校验。其中,所述一级校验用于确定所述待监测内容是否存在异常。
本实施例中,一级校验可以理解为是初校验,其主要任务是初步校验待监测内容是否合规。其中,一级校验的过程可以如上述步骤S203的过程所述,即可以确定第一内容标识码是否残缺,在残缺的情况下,可以确定待监测内容存在异常;在未残缺的情况下,可以继续确定第一内容标识码是否乱码,若出现乱码,则一级校验不通过,表征待监测内容存在异常;若未出现乱码,则可以继续解析出第一内容标识码所标识的内容属性,基于内容属性,确定该待监测内容是否是在该待监测的播放时段,指定该显示设备应该播放 的内容,若确定该待监测内容并非在该待监测的播放时段应该播放的内容,或者该待监测内容对应的显示设备的标识,并非是当前的显示设备,则可以确定待监测内容是异常的,一级校验不通过;反之,则是合规的,一级校验通过。
本可选示例中,若一级校验通过,即表征待监测内容是合规的情况下,可以结束监测,不进入到S33;反之,若一级校验不通过,即表征待监测内容是不合规的情况下,继续进入S33,进行二级校验,以排查异常原因后,再最终判定待监测内容是否合规。
S33,对显示页面截图进行二级校验,二级校验用于确定待监测内容的异常类别,并根据二级校验的结果,确定待监测内容是否合规。
本可选示例中,二级校验主要用于排查待监测内容出现异常的异常类别,该异常类别即可以理解为是导致待监测内容出现异常的原因,实际中,其中,显示页面截图由于是对显示设备的当前屏幕进行截图得到的,其不仅携带待监测内容的显示画面和第一内容标识码,还携带了当前屏幕上的其他显示画面,信息丰富,因此可以基于显示页面截图进行异常的归因分析,之后,再基于分析出的异常类别,确定待监测内容是否合规。
具体地,二级校验即为上述的合规性判定阶段,具体而言,在一种可选的示例中,由于异常的原因可以包括屏幕显示故障的原因、第三方入侵的原因,而这两种原因都可以基于对携带丰富信息的显示页面截图得到。
在一种可选的示例中,可以对显示页面截图进行画面异常检测,确定待监测内容是否存在第一类型的异常画面和/或第二类型的异常画面;第一类型为屏幕异常的类型,第二类型为包含内容异常的类型。
相应地,若存在第一类型的异常画面,则确定待监测内容合规;若存在第二类型的异常画面,则确定待监测内容不合规。
本可选示例中,第一类型是屏幕异常的类型,该屏幕异常可以是指:显示设备的显示屏出现了显示异常,属于显示设备的软硬件故障,具体而言,可以通过检测显示页面截图中是否具有黑屏和/或花屏的画面,确定待监测内容的异常是否归因于屏幕异常。
具体来说,若显示页面截图中具有黑屏的画面,则可以确定待监测内容的异常归因于屏幕异常;若显示页面截图中具有花屏的画面,则可以在花屏 的画面的数量大于预设数量的情况下,确定待监测内容的异常归因于屏幕异常;若显示页面截图中具有黑屏和花屏的画面,则确定待监测内容的异常归因于屏幕异常。其中,若花屏的画面的数量不大于预设数量,则可以结合第二类型的异常画面检测的结果,再确定待监测内容的异常归因。
本可选示例中,第二类型是内容异常的类型,实际中,可以基于显示页面截图中除第一内容标识码外的图像区域中的画面,确定待监测内容是否存在内容异常,其中,内容异常可以是指:具有涉黄涉毒或者其他一些敏感内容的画面。具体实施时,可以基于机器学习的手段实现对内容异常的检测。
如图3所示,在二级校验中,可以先检测显示页面截图中具有黑屏和/或花屏的画面,以确定待监测内容的异常是否是屏幕异常导致的,若是,则可以确定异常是显示设备的软硬件故障导致的,不属于在播内容的内容异常,因此可以确定待监测内容是合规的,如此可以结束校验,并发送报警信号,以指示对显示故障的软硬件故障进行修复。
若显示页面截图中不具有黑屏的画面,或者具有少量的花屏画面,则需要再检测显示页面截图中是否具有涉黄涉毒或者其他一些敏感内容等内容异常的画面,若检测不存在内容异常的画面,则确定待监测内容是合规的;若检测存在内容异常的画面,则确定待监测内容是不合规的,如此,可以强制关闭显示设备,并发送报警信号,以指示增强媒体资源管理***应对第三方入侵的安全防护。
在又一种可选的示例中,在确定待监测内容是否存在第一类型的异常画面中,以及在确定待监测内容是否存在第二类型的异常画面中,均可以基于机器学习的手段,得到监测结果。
具体而言,机器学习的手段可以是指预先利用训练样本训练出对应的检测模型,以通过检测模型对显示页面截图进行分析,从而得到待监测内容是否存在第一类型和第二类型的异常画面的结果。
具体地,可以将显示页面截图输入至第一检测模型,并根据第一检测模型输出的结果确定待监测内容是否存在所述第一类型的异常画面。和/或,将显示页面截图输入至第二检测模型,并根据第二检测模型输出的结果确定待监测内容是否存在第二类型的异常画面。其中,第一检测模型是以多个携带有屏幕异常的图像为训练样本对第一神经网络进行训练得到的,第二检测模 型是以多个携带异常内容的图像为训练样本对第二神经网络进行训练得到的。
本实施例中,可以先将显示页面截图输入至第一检测模型,以确定待监测内容是否存在第一类型的异常画面。其中,第一检测模型输出的结果可以是两个概率值,分别对应存在第一类型的异常画面的类别,以及不存在第一类型的异常画面的类别。实际中,可以选择较大的概率值对应的类别作为检测结果,示例地,概率值分别为0.8和0.2,0.8是对应存在第一类型的异常画面的类别,则可以确定检测结果是待监测内容存在第一类型的异常画面。
实际中,若第一检测模型输出的结果表示待监测内容不存在第一类型的异常画面,则可以继续将显示页面截图输入至第二检测模型,并根据第二检测模型输出的结果确定待监测内容是否存在第二类型的异常画面。同理,第二检测模型输出的结果也可以是两个概率值,分别对应存在第二类型的异常画面的类别,以及不存在第二类型的异常画面的类别。实际中,可以选择较大的概率值对应的类别作为检测结果,示例地,概率值分别为0.7和0.4,0.7是对应存在第二类型的异常画面的类别,则可以确定检测结果是待监测内容存在第二类型的异常画面。
当然,若第一检测模型输出的结果表示待监测内容存在第一类型的异常画面,也可以继续将显示页面截图输入至第二检测模型,继续确定待监测内容是否存在第二类型的异常画面,从而第二检测模型输出的结果,确定依据待监测内容是否存在第二类型的异常画面,若存在第二类型的异常画面,判定待监测内容不合规,若不存在第二类型的异常画面,则判定待监测内容合规但是判定显示设备异常。
其中,第一检测模型的训练样本可以是包括黑屏和花屏的样本图像,第二检测模型可以依据待检出的异常内容,收集包含涉黄涉毒或者其他一些敏感内容的图像作为训练样本,其中,采用训练样本进行训练,得到第一检测模型和第二检测模型的过程可以采用相关技术的内容,在此不再赘述。
其中,第一检测模型可以为常用的分类模型,例如可以是任一一种模型:VGG19、Resnet512、inceptionV3;第二检测模型也可以为常用的分类模型,例如VGG19、Resnet512、inceptionV3中的任一种;其中,第一检测模型和第二检测模型可以是同一种类别的模型,如都是VGG19,也可以是不同类型的模型。
由于利用训练样本训练出第一检测模型以对屏幕异常进行检测,以及利用训练样本训练出第二检测模型以对内容异常进行检测,由此可以依靠模型的泛化性能,提高对待监测内容的合规性监测的准确性。
如图1所述,本公开的播放内容的监测方法可以由播控主机执行,则在由播控主机执行的情况下,其中,一级校验可以由播控主机进行;二级校验则可以由内容发布服务器执行。
相应地,在一级校验不通过的情况下,即检测到待监测内容存在异常的情况下,播控主机可以将显示页面截图发送给内容发布服务器,以使内容发布服务器进行二级校验;之后,播控主机可以接收内容发布服务器返回的二级校验的结果,并基于该二级校验的结果,确定待监测内容是否合规。
其中,内容发布服务器进行二级校验的过程可以参照上述S33的描述过程,在此不再赘述。当内容发布服务器得到二级校验的结果时,可以将该二级校验的结果发送给播控主机,进而播控主机基于二级校验的结果,确定待监测内容是否合规。如上所述,在二级校验的结果表示存在第二类型的异常画面,判定待监测内容不合规,播控主机强制关闭显示设备;若表示不存在第二类型的异常画面,则判定待监测内容合规。
当然,在又一种可选的示例中,内容发布服务器得到二级校验的结果后,也可以不将结果告知播控主机,而直接输出待监测内容是否合规的信息,在待监测内容不合规时,直接向播控主机下达关闭显示的指令,从而强制关闭显示设备。
采用本公开实施例的监测机制,具有以下优点:
一方面,由于采用了一级校验和二级校验的分阶段监测机制,在以第一内容标识码为依据的一级校验不通过时,再进行以显示页面截图为依据的二级校验,从而使得一级校验和二级校验依据了不同数据量的信息,由此,无需每次对待监测内容进行监测时,都进行全局(显示页面截图和第一内容标识码)的分析,而是先基于数据量小的第一内容标识码进行监测,不通过时,再基于数据量丰富的显示页面截图进行分析,由此在提高监测效率和监测准确度的同时,避免长时间高占比地消耗计算资源,从而提高了计算资源的利用率。
再一方面,由于一级校验由播控主机进行,二级校验由内容发布服务器 进行,从而将数据量较小的一级校验放在播控主机上进行,由于播控主机与显示设备直接连接,在媒体资源管理***中的播控主机一般有多个,一个播控主机连接一个或多个显示设备,其属于边缘的设备,可以达到利用多个边缘端分散性地对显示设备的在播内容的高效监测,且由于分散性检测,也避免了网络并发的压力。而一级校验不通过时,才由内容发布服务器进行数据量较大的二级校验,这样可以合理利用内容发布服务器的优势计算资源准确性进行合规性判定;这样,采用采用分级响应的策略,有助于媒体资源管理***快速作出响应,同时减少网络的并发压力,降低内容发布服务器的计算压力,提高媒体资源管理***的监测效率。
2.1,确定待监测内容是否合规过程中的信息处理
2.1.1,对第一内容标识码和第二内容标识码的处理
由于内容标识码可以唯一标识播放的内容,因此,在从显示设备获取到显示页面的显示页面截图时,显示页面截图中包括在播内容的画面,如上述实施例所述,由于除需要及时监测显示设备是否黑屏、花屏外,还需要确定在播内容是否应该在当前显示设备播放,以及在播内容的播放时间段是否正确等,因此,内容标识码可以用于标识播放内容的名称、播放时段、所播放的显示设备等属性。
其中,在一个显示设备上,不同的播放时段可能播放不同的内容,这样,在基于第一内容标识码,对显示设备的待监测内容进行监测时,还可以以指定在当前的播放时段进行播放的预订内容所对应的第二内容标识码作为参照,即作为第一内容标识码的对照,验证待监测内容是否是在当前的播放时段指定播放的预订内容。
具体实施时,在每向显示设备发送待播放的内容之前,可以生成待播放的内容的身份编码信息,该身份编码信息可以包括播放内容的名称、播放时段、所播放的显示设备等信息的编码。由此,可以基于身份编码信息得到每个待播放的内容的第二内容标识码,并将第二内容标识码存储到本地,以便后续基于第一内容标识码和该第二内容标识码,确定待监测内容是否合规。
参照图4所示,示出了本公开的一种播放内容的监测方法的整体流程示意图,如图4所示,具体包括以下流程:
S41:在每向显示设备发送待播放的内容之前,可以生成待播放的内容对应的第二内容标识码,其中待播放的内容可以理解为是指定显示设备播放的预定内容。
具体地,可以对待发送给显示设备进行播放的预定内容,生成与预定内容对应的身份编码信息;之后,基于身份编码信息,生成预定内容的第二内容标识码。
本可选实施例中,身份编码信息可以是对内容的名称、内容标识、播放时段和显示设备的标识等进行编码后得到的具有指定位数的编码,其可以理解为是预定内容的身份ID,该身份编码信息可以与预定内容以及播放预定内容的显示设备相关联。
一般而言,身份编码信息可以具有指定的位数,以反映预定内容的多维度的内容属性;在一种示例中,身份编码信息可以是对对内容的名称、内容标识、播放时段和显示设备的标识等进行三十二进制的编码后得到的,也就是说身份编码信息可以是三十二进制的信息,当然,在其他一些示例中,身份编码信息也可以是十六进制的信息;具体可以依据实际要编码的内容属性的类型和多寡确定。例如,身份编码信息可以是26位的三十二进制的身份信息,具体的编码过程和内容标识码的生成过程在后续2.1.2节详述,此处不做赘述。
其中,可以将身份编码信息转换为第二内容标识码,转换过程可以是将身份编码信息进行加密后得到第二内容标识码,或者第二内容标识码可以是将身份编码信息加密后再进行转换得到的二维码,实际中,只要可以将身份编码信息进行转换得到第二内容标识码即可,具体的第二内容标识码采用哪一格式在此不做限制,具体可以参见后续2.1.2节的详述。
S42:在生成第二内容标识码后,可以将待播放的预订内容发送给对应的显示设备进行播放,在将待播放的预订内容发送给对应的显示设备时,还可以将第二内容标识码发送给显示设备,以使显示设备在播放预定内容时,在显示页面展示第二内容标识码。其中,在显示设备播放预订内容时,可以将第二内容标识码显示在屏幕上的固定位置,如上所述,可以展示在屏幕的左上角位置、右上角位置、左下角位置或右下角位置,也可以展示在播放窗口的左上角位置、右上角位置、左下角位置或右下角位置。优选地,为了不 影响观看者对预定内容的观看,可以将第二内容标识码展示在屏幕的四角。
S43:以存储的第二内容标识码为第一内容标识码的对照,确定待监测内容是否合规。
其中,可以将预订内容对应的第二标识码与预订内容对应的播放时段进行关联存储,这样,在一个播放时段中,获取到显示设备的当前播放的待监测内容对应的第一内容标识码时,可以以存储的与该播放时段对应的第二内容标识码为第一内容标识码的对照,以确定待监测内容是否合规。
相应地,在基于第一内容标识码,确定待监测内容是否合规时,可以获取与待监测的播放时段对应的第二内容标识码,其中,第二内容标识码与在待监测的播放时段进行播放的预定内容对应;基于第一内容标识码和第二内容标识码,确定待监测内容是否合规。
其中,在以第二内容标识码为参照确定待监测内容是否合规时,在一种可选示例中,可以直接基于第一内容标识码和第二内容标识码,确定待监测内容是否合规,即可以不进行后续的二级校验,单独依据一级校验结果便确定待监测内容是否合规。具体实施时,可以比对第一内容标识码和第二内容标识码是否一致,若一致,表示显示设备在待监测的播放时段播放的是预定内容,则待监测内容是合规的;若不一致,则表示显示设备在待监测的播放时段播放的可能不是预定内容,则待监测内容是不合规的。
当然,在又一种可选示例中,为了避免误检,仍然可以采用一级校验和二级校验相结合的监测机制,即在比对第一内容标识码和第二内容标识码不一致的情况下,可以基于显示页面截图进行二级校验,以判定待监测内容是否合规。
其中,在该可选的实施例中,可以先基于第一内容标识码和第二内容标识码各自对应的第一中间信息进行监测,具体而言是基于各自对应的第一中间信息,确定第一内容标识码是否与第二内容标识码一致,并基于是否一致的结果,确定待监测内容是否合规。
S44:基于第二内容标识码和第一内容标识码各自对应的第一中间信息的校验。
相应地,可以别对第二内容标识码和第一内容标识码进行解析,得到第二内容标识码和第一内容标识码分别对应的第一中间信息;并基于第二内容 标识码和第一内容标识码分别对应的第一中间信息,确定待监测内容是否合规;其中,第一中间信息为所播放的内容的身份编码信息的摘要。
其中,在基于第二内容标识码和第一内容标识码分别对应的第一中间信息,确定待监测内容是否合规时,可以比对第二内容标识码的第一中间信息和第一内容标识码的第一中间信息是否一致,若一致,则待监测内容是合规的,若不一致,则待监测内容是不合规的。
本实施例中,第一中间信息是身份编码信息在转换为内容标识码的过程中的中间信息,可以理解为是身份编码信息的摘要,如身份编码信息经过第一次转换、第二次转换甚至第三次及更多次转换后,得到内容标识码,则第一中间信息可以是第二次、第三次转换得到的中间信息。这样,先比对第一中间信息,可以避免对对第二内容标识码和第一内容标识码进行多次反转换,从而提高监测效率。
S45:基于第二内容标识码和第一内容标识码各自对应的第二中间信息的校验。
当然,若第二内容标识码和第一内容标识码分别对应的第一中间信息不完全一致,则为了监测准确性,还可以基于内容标识码进行更进一步的验证。
具体地,第一内容标识码和第二内容标识码还包括隐藏区,隐藏区用于携带身份编码信息对应的第二中间信息,则可以确定第二内容标识码和第一内容标识码分别对应的第二中间信息是否完全一致。其中,第二中间信息和第一中间信息都可以是在将身份编码信息转换为内容标识码的过程中的中间产物,第二中间信息可以早于第一中间信息产出,如此,可以理解为是第一中间信息是基于第二中间信息生成的,由此,第二中间信息可以携带更多的原始信息,或者可以理解为第二中间信息更接近原始的身份编码信息,由此,再次基于第二中间信息可以准确判定待监测内容是否合规。其中:
若第二内容标识码和第一内容标识码分别对应的第一中间信息完全一致,则确定待监测内容合规,结束流程。
若第二内容标识码和第一内容标识码分别对应的第一中间信息不完全一致,则分别对第二内容标识码和第一内容标识码的隐藏区进行解析,得到对第二内容标识码和第一内容标识码各自对应的第二中间信息;并基于第二内容标识码和第一内容标识码各自对应的第二中间信息,确定待监测内容是否 合规。
具体实施时,可以按位比对两个第二中间信息;其中,在两个第二中间信息完全一致的情况下,确定待监测内容合规;在两个第二中间信息不完全一致的情况下,可以进行二级校验,或者直接认定为不合规,结束流程。
在又一种可选的示例中,在基于更接近身份编码信息的第二中间信息,确定待监测内容是否合规时,还可以基于第一内容标识码与第二内容标识码不一致的字段,生成对应等级的报警信息,以指示何种类型的不一致,如指示是播放时段的不一致,还是播放的显示设备的不一致。
在该可选示例中,身份编码信息的不同字段可以是对预定内容的不同内容属性进行编码得到的,也就是说不同字段表征了预定内容的不同属性,如名称属性、播放时段属性和播放设备的设备标识属性都可以在不同字段中予以体现。
相应地,在两个第二中间信息不完全一致的情况下,可以获取两个第二中间信息不一致的字段,以及不一致的字段所标识的内容属性;基于不一致的字段所标识的内容属性,生成对应等级的报警信号;其中,内容属性包括显示设备的标识属性、播放时段属性中的至少一者。
本实施例中,在两个第二中间信息不完全一致的情况下,便可以依据不一致的字段对应的内容属性,生成对应等级的报警信息,具体而言,不同的内容属性可以对应不同等级的报警,例如,可以将显示设备的标识属性设为低级别的报警,播放时段属性设置为高级别的报警,实际中,还可以将内容的名称设置为中等级别的报警等,当然,此处仅为示例性说明,不代表对本公开的限制。
如此,在报警信号是文本类的信息的情况下,该报警信号可以携带不一致的字段对应的内容属性,以指示显示设备在哪一方面出现不合规的播放。
当然,作为一种补充实施例,也可以将第一内容标识码直接转换为原始编码的身份编码信息,这样,可以基于第一内容标识码和第二内容标识码各自对应的身份编码信息,确定待监测内容是否合规,也就是依据原始的身份编码信息进行校验。
其中,在依据原始的身份编码信息进行校验的情况下,第一内容标识码对应的身份编码信息,可以直接从隐藏区提取第二中间信息,之后将该二进 制的第二中间信息进行三十二进制转换后,得到原始的身份编码信息。
采用此种实施方式时,由于可以依据第二内容标识码作为对照,以更加细粒度地从播放时段、显示设备等维度监测显示设备的在播内容的合规性,由此,可以应用于有广告投放业务的媒体资源管理***中,以准确监控广告内容的播放时段是否正确,播放其广告内容的显示设备是否正确。同时,还可以基于一级校验和二级校验相结合的方式,及时检出被第三方入侵而播放的非法内容,提高了内容监测的效率,且拓宽了内容监测的应用场景。
2.1.2,内容标识码的生成过程
如上所述,内容标识码是需要在显示设备显示的预定内容对应的标识码,为与正在显示设备上播放的内容对应的第一内容标识码区分,该预定内容对应的标识码称为第二内容标识码。
在一种可选示例中,内容标识码可以是图像类型的标识码,该内容标识码可以通过图像上的颜色信息隐藏在播内容的身份编码信息。也就是说,在一种情况中,可以将文本类型的身份编码信息转换为图像类型的信息,以通过图像去唯一表征文本内容,此种方式可以理解为是信息的类转换,即将信息从一个格式类别转换为另一个格式类别,而信息的含义保持不变,从而达到对信息的高级别的保密,避免信息的泄露,从而维护信息的安全。
参照图5所示,示出了本公开的将文本类型的身份编码信息转换为图像类型的内容标识码的原理示意图,结合图5所示,对本公开的第二内容标识码的生成过程进行说明。
首先,需要生成身份编码信息,本实施例中,身份编码信息可以是对预订内容的多维度的内容属性分别进行编码得到的。具体实施时,可以获取预定内容的多维度的内容属性;对多维度的内容属性进行三十二进制编码,得到第二预定位数的身份编码信息。其中,多维度的内容属性包括所述预定内容的名称、播放时段以及播放预定内容的显示设备的设备标识中的一者或多者。
参照图6所示,示出了一种示例中身份编码信息的编码示意图,如图6所示,身份信息编码可以是26位的三十二进制的编码,起始0-5位表示内容的名称;第6位表示该内容的播放类型(图6中的计划类型),类型包含正常播放类型、垫片类型、黑屏;第7位表示屏幕被分割的窗口类型,窗口类 型可以事先定义并编号,也可以在做计划时自定义并编号,举例来说,“1”代表一个窗口;“2”代表两个窗口,横向分布;“3”代表两个窗口,纵向分布;“4”代表三个窗口,一行,横向分布;“5”代表三个窗口,两行,第一行两个窗口,第二行一个窗口;“6”代表三个窗口,两行,第一行一个窗口,第二行两个窗口;“7”代表三个窗口,三行,纵向分布。
其中,身份编码信息中的第8位为内容的节目类型,具体有广告视频、广告海报、公交路线图、公交车时刻表、通知公告文本等等可在媒体终端显示的信息。第9-14位表示内容的节目ID;第15-17位表示待播放的显示设备的设备ID,第18-21位表示内容要开始播放的起始时间;第22-25位表示内容要接收播放的结束时间。
其中,在将身份编码信息转换为内容标识码时,如图5所示,可以选用以下两种可选示例进行。
可选示例1:将身份编码信息对应的第一中间信息中每位的值映射为透明图像中对应校验点的灰度值,得到内容标识码。
可选示例2:在可选示例1得到的内容标识码的基础上,在透明图像中新增隐藏区,将身份编码信息对应的第二中间信息映射到隐藏区,从而得到携带两种中间信息的内容标识码。
2.1.2.1,对可选示例1。
具体实施时,可以将对所述身份编码信息迭代进行多种编码,得到每种编码下的中间信息;将多种编码各自对应的中间信息中的第一中间信息映射为预定透明图像上的多个校验点,得到第二内容标识码。
其中,第一中间信息为十进制的灰度信息,第一中间信息中不同位置的数值对应不同位置的校验点。
参照图5所示,可以对身份编码信息迭代先进行二进制转换;再将二进制转换得到的中间信息进行十六进制转换;并将十六进制转换得到的中间信息进行十进制转换,得到十进制的灰度信息。
其中,十进制的灰度信息即为第一中间信息,其中,第一中间信息可以具有预定位数,如是32位的十进制信息,这样第一中间信息便具有32个数值,每个数值的值在0-9之间。其中,在预定透明图像上可以包含多个校验点,校验点的数量与第一中间信息的为数可以是相同的,如是32个校验点。
这样,可以将第一中间信息中每位的数值作为对应位置的校验点的灰度值,从而为预定透明图像上的校验点赋予了灰度,此种情况下,第二内容标识码便是具有多个校验点,且校验点具有灰度而其他区域透明的图像。
其中,由于第一中间信息中不同位置的灰度值对应不同位置的校验点,实际中可以指定第一中间信息中各位数值与校验点的位置对应关系,示例地,第一中间信息从左到右的数值,依次对应预设透明图像上的从上到下和从左到右的校验点,对应时,先按照从上到下的位置关系找到一行的校验点,再按照从左到右的顺序找到该行中的对应校验点。
当然,预设透明图像的多个校验点在预设透明图像上可以是阵列排布,也可以是以列排布或者可以是一行排布,或者也可以是其余形状的排布,在此不做限定。
其中,校验点可以是圆形的点、或者是矩形的点,或者可以是其他形状的点,在此不做限制。
此种示例中,在基于第一内容标识码和第二内容标识码,确定待监测内容是否合规时,可以分别对第一内容标识码和第二内容标识码中的校验点的灰度值进行解析,之后,按照校验点与位之间的位置对应关系,将各个灰度值组合,从而反解析出第一中间信息,从而如上所述,基于两个第一中间信息,判断待监测内容是否合规。
2.1.2.2,对可选示例2。
如上述实施例所述,由于内容标识码可以携带信息隐藏区,则在将文本类的身份编码信息转换为图像类的内容标识码时,可以在预定透明图像中设置一个隐藏区,之后,利用信息隐藏区携带身份编码信息对应的第二中间信息。
其中,如图5所示,隐藏区可以设置在预定透明图像的下方,当然图5仅是示例性说明,实际中,隐藏区可以设置在预定透明图像的上方、左侧或右侧。
相应地,可以获取多种编码各自对应的中间信息中的第二中间信息;第二中间信息早于第一中间信息产出,第二中间信息为二进制的信息;并基于第二中间信息中每位的编码值,更新信息隐藏区的灰度值。
本实施例中,第二中间信息可以是将身份编码信息进行二进制转换后得 到的信息,例如,可以是将身份编码信息将使用ASCII编码,从而使得每个字符包含8位二进制编码,由此,可以将二进制编码信息中每位的0、1作为隐藏区中一个小区间的灰度,从而得到隐藏区。其中,小区间可以是将隐藏区进行按照第二中间信息的位数进行划分后得到的,不同小区间对应第二中间信息的不同位置的值。
示例地,由于身份编码信息是三十二进制的,包含26位数值,则在将身份编码信息将使用ASCII编码(American Standard Code for Information Interchange)后,每位字符包含8位二进制编码,则第二中间信息具有208位,每位的值要么是0要么是1,则可以将208位的值依次作为隐藏区的对应小区间的灰度值。
此种情况下,在基于第一内容标识码和第二内容标识码分别对应的第二中间信息,确定待监测内容是否合规时,可以分别对第一内容标识码和第二内容标识码中的隐藏区的每个小区间的灰度值进行解析,之后,按照小区间与位之间的位置对应关系,将各个灰度值组合,从而反解析出第二中间信息,从而如上所述,基于两个第二中间信息,判断待监测内容是否合规。
采用本公开实施例的信息处理手段,具有以下优点:
第一,由于将身份编码信息进行多种编码转换后,得到内容标识码,这样,可以提高内容标识码的安全性,使得内容标识码不易被第三方破解,从而增强了对媒体资源管理***巡检的安全防护性。
第二,由于将身份编码信息转换为了图像类别的内容标识码,由此将信息从一个格式类别转换为另一个格式类别,而信息的含义保持不变,从而达到对信息的高级别的保密,进一步增加了内容标识码的破解难度。
第三,由于将身份编码信息被转换过程中的中间信息隐藏至隐藏区,在进行校验时,先是依据第一中间信息进行校验,由于第一中间信息只需进行很少次的反解析,因此提高了监测速度;之后,依据第一中间信息校验不通过时,再依据第二中间信息校验,由于第二中间信息更接近原始的身份编码信息,因此包含足够的信息量,这样依据第二中间信息的校验可以提高监测准确率,在一级校验阶段可以准确检出待监测内容的播放异常,从而在兼顾监测速度的同时,保证了监测准确率。
当然,实际中,在依据第二中间信息反解出原始的身份编码信息后,可 以依据第一内容标识码和第二内容标识码各自对应的身份编码信息进行校验,由于依据的是原始的身份编码信息,因此可以得到最准确的校验结果。
下面,以图1所示的通信环境,结合一个具体示例,对本公开的播放内容的监测方法进行示例性说明。该示例以一级校验和二级校验对待监测内容进行合规性监测为例说明,参照图7所示,示出了本示例的一级校验的完整流程示意图,参照图8所示,示出了本示例的二级校验的完整流程示意图。
其中,如图7所示,一级校验的完整流程如下:
S71:内容发布服务器为每个内容生成对应的内容标识码。
将内容的名称、播放设备、播放设备、播放起始时间、播放结束时间、播放类型以及内容的类别编码为图6所示的身份编码信息;
接着,将身份编码信息使用ASCII编码,每个字符包括8位二进制编码,得到二进制的中间信息(第二中间信息);对二进制的中间信息使用MD5或者SHA1编码,生成32位的十六进制信息(第一中间信息),将十六进制信息转成十进制的灰度信息,将灰度信息中每位的值作为预定透明图像上对应校验点的灰度值;将第二中间信息的每位数值作为灰度,放置在信息隐藏区。
S72:以图1的LED显示屏作为显示设备为例,播控主机将预定内容1和对应的内容标识码发送给LED显示屏显示,以使LED显示屏在播放预定内容1时,显示内容标识码。
S73:在预定内容1的播放时段,播控主机从LED显示屏获取显示页面截图,从而显示页面截图中提取第一内容标识码,以及从本地获取该播放时段对应的第二内容标识码。
S74:播控主机从第一内容标识码中解析出各个校验点的灰度值,然后按照位置对应关系组合灰度值后,得到的是十进制的信息,之后,将十进制的信息转换为十六进制的信息(第一内容标识码对应的第一中间信息);同理,可以直接获取预定内容的身份编码在转换过程中存储的十六进制信息(第二内容标识码对应的第一中间信息)。
S75:比对两个十六进制的信息是否一致,若一致,结束流程,并输出成功代码1,上传内容发布服务器该显示页面截图以留底;若不一致,则继续从第一内容标识码的隐藏区中解析出二进制的中间信息(第二中间信息), 将该二进制的中间信息转换回三十二进制的身份编码信息。
S76:比对两个身份编码信息是否一致,若完全一致,则结束流程,并输出成功代码2,上传内容发布服务器该显示页面截图以留底;若不一致,则按位比对两个身份编码信息。
S77:比对显示设备的设备标识所对应的字段是否一致,即比对设备标识的内容属性是否一致,若一致,则输出报警代码1,上传显示页面截图给内容发布服务器,进行二级校验;若不一致,则继续比对播放时段对应的字段是否一致,即比对播放时段的内容属性是否一致,若一致,则输出报警代码2,上传显示页面截图给内容发布服务器,进行二级校验;若不一致,则输出报警代码3,上传显示页面截图给内容发布服务器,进行二级校验。
其中,报警代码1的级别低于报警代码2的级别,报警代码2的级别低于报警代码3的级别。
其中,如图8所示,二级校验的完整流程如下:
S81:播控主机将显示页面截图和对应的报警代码发送给内容发布服务器;
S82:内容发布服务器在检测到是警告代码1和报警代码2的情况下,直接报告管理员,让其确认待监测内容是否是在正确的显示设备上,于正确的播放时段播放。
S83:内容发布服务器在检测到是警告代码3的情况下,首先进行黑屏和花屏等播放不全的屏幕异常检测,例如,将显示页面截图输入到第一检测模型,如果第一检测模型输出的结果是屏幕黑屏/花屏/播放不全等,与预计情况不符,则将信息上报给管理员,对LED显示屏进行维护。
S84:排除黑屏、花屏等屏幕异常后,进行涉黄涉爆和敏感信息检测,例如,将显示页面截图输入到第二检测模型,第二检测模型可以使用计算机视觉算法和OCR等技术提取关键字,如果第二检测模型输出的结果是具有上述非法内容,则指示播控主机直接关闭LED显示屏,并向管理员报警。
S85:排除以上异常(黑屏、花屏、非法内容)以后,则获取待监测内容的类别,以留档并上报管理员。
总体上而言,采用本公开的播放内容的监测方法,对媒体资源管理***分发至多种远端屏幕(LCD显示屏、LED显示屏以及其他类型的显示终端上 的显示屏)的播放内容的安全性进行分级监控。通过部署在边缘端的播控主机首先进行一级校验,若一级校验通过,则直接作出相应处理,并将处理结果内容服务器留底。若一级校验不通过,则将显示页面截图反馈至内容服务器进行更智能的二级校验,以便作出相应决策。采用边缘计算和分级校验的策略,可以提高边缘端设备(播控主机)的使用率。有助于监控***快速作出响应,同时减少网络的并发和丢包影响,降低内容发布服务器的计算压力。
毫无疑问的,本公开所公布的加密算法SDK和校验算法SDK,可以部署在任意设备的处理器或存储器中。媒体资源管理***分发也可发送至手机、平板电脑、笔记本电脑等应用终端。进而,也可以对媒体资源管理***分发只至手机、平板电脑、笔记本电脑等终端的播放内容进行合规性检测。
基于相同的发明构思,本公开还提供了一种播控主机,所述播控主机用于执行所述的播放内容的监测方法。
如图1所示,播控主机可以是部署在媒体资源管理***中的边缘端,直接与各种类型的显示设备连接,一个播控主机可以连接一个或多个显示设备。
基于相同的发明构思,本公开还提供了一种监播***,参照图9所示,示出了一种监播***的***结构示意图,如图9所示,包括:内容发布服务器,与所述内容发布服务器连接的至少一个播控主机,以及与每个所述播控主机连接的显示设备;
所述内容发布服务器,用于将待播放的内容发送给对应的播控主机;
所述播控主机,用于将所述待播放的内容发送给所述显示设备,并用于执行所述的播放内容的监测方法;
所述显示设备,用于播放所述待播放的内容。
本可选示例中,内容发布服务器可以是内容发布平台的后台服务器,内容发布平台可以是web端应用,也可以是移动端应用,如APP。
在一种可选示例中,播控主机可以执行上述实施例中所述的一级校验和二级校验,在又一种可选示例中,播控主机可以执行上述实施例中所述的一级校验,内容发布服务器可以执行上述实施例中的二级校验。
其中,图9中示出了三个播控主机,五个显示设备的情况,实际中,播控主机的数量和显示设备的数量可以是不受图9的限制。
基于相同的发明构思,本公开还提供了一种播放内容的监测装置,参照 图10所示,示出了本公开实施例的监测装置的结构示意图,如图10所示,所述装置包括:
截图获取模块1001,用于在当前的监测时段,从显示设备获取显示页面截图,所述显示页面截图是对所述显示设备当前播放的待监测内容进行截图得到的;
标识码获取模块1002,用于从所述显示页面截图中提取第一内容标识码,所述第一内容标识码用于标识所播放的内容;
验证模块1003,用于基于所述第一内容标识码,确定所述待监测内容是否合规。
可选地,所述验证模块1003,包括:
第一验证单元,用于基于所述第一内容标识码,进行一级校验,所述一级校验用于确定所述待监测内容是否存在异常;
第二验证单元,用于在所述一级校验不通过的情况下,对所述显示页面截图进行二级校验,所述二级校验用于确定所述待监测内容的异常类别;
判断单元,用于基于所述二级校验的结果,确定所述待监测内容是否合规。
可选地,所述第二验证单元,具体用于对所述显示页面截图进行画面异常检测,确定所述待监测内容是否存在第一类型的异常画面和/或第二类型的异常画面;第一类型为屏幕异常的类型,第二类型为包含内容异常的类型;
所述判断单元,具体用于若存在所述第一类型的异常画面,则确定所述待监测内容合规;若存在第二类型的异常画面,则确定所述待监测内容不合规。
可选地,所述第二验证单元,包括:
第一检测子单元,用于将所述显示页面截图输入至第一检测模型,并根据所述第一检测模型输出的结果确定所述待监测内容是否存在所述第一类型的异常画面;
和/或,
第二检测子单元,用于将所述显示页面截图输入至第二检测模型,并根据所述第二检测模型输出的结果确定所述待监测内容是否存在第二类型的异常画面;其中,所述第一检测模型是以多个携带有屏幕异常的图像为训练样 本对第一神经网络进行训练得到的,所述第二检测模型是以多个携带异常内容的图像为训练样本对第二神经网络进行训练得到的。
可选地,所述第二验证单元,包括:
发送子单元,用于将所述显示页面截图发送给内容发布服务器,以使所述内容发布服务器进行所述二级校验;
所述判定单元,具体用于基于所述内容发布服务器返回的二级校验的结果,确定所述待监测内容是否合规。
可选地,所述装置还包括以下模块:
关闭模块,用于在确定所述待监测内容不合规的情况下,关闭所述显示设备,以禁止所述显示设备进行内容播放;
和/或,报警模块,用于在确定所述待监测内容不合规的情况下,发送报警信号,以指示所述待监测内容不合规。
可选地,所述装置还包括:
编码信息生成模块,用于对待发送给所述显示设备进行播放的预定内容,生成与所述预定内容对应的身份编码信息;
内容标识码生成模块,用于基于所述身份编码信息,生成所述预定内容的第二内容标识码;
发送模块,用于将所述预定内容与所述第二内容标识码发送给所述显示设备,以使所述显示设备在播放所述预定内容时,在显示页面展示所述第二内容标识码;其中,所述第二内容标识码作为所述第一内容标识码的对照,以用于确定所述待监测内容是否合规。
可选地,所述验证模块1003,包括:
第二内容标识码获取单元,用于获取与所述待监测的播放时段对应的第二内容标识码,其中,所述第二内容标识码与在所述待监测的播放时段进行播放的预定内容对应;
验证单元,用于基于所述第一内容标识码和所述第二内容标识码,确定所述待监测内容是否合规。
可选地,所述验证模块1003,包括:
解析单元,用于分别对所述第二内容标识码和所述第一内容标识码进行解析,得到所述第二内容标识码和所述第一内容标识码分别对应的第一中间 信息;其中,所述第一中间信息是将内容对应的身份编码信息转换为内容标识码的过程中的中间信息;
第一判断单元,用于基于所述第二内容标识码和所述第一内容标识码分别对应的第一中间信息,确定所述待监测内容是否合规;
其中,所述第一中间信息为所播放的内容的身份编码信息的摘要。
可选地,所述第二内容标识码还包括隐藏区,所述隐藏区用于携带身份编码信息对应的第二中间信息,所述第一判断单元,具体用于:
确定所述第二内容标识码和所述第一内容标识码分别对应的第一中间信息是否完全一致;
若否,则分别对所述第二内容标识码和所述第一内容标识码的隐藏区进行解析,得到对所述第二内容标识码和所述第一内容标识码各自对应的第二中间信息;其中,所述第二中间信息在所述第一中间信息之前产出;
基于所述第二内容标识码和所述第一内容标识码各自对应的第二中间信息,确定所述待监测内容是否合规。
可选地,基于所述第二内容标识码和所述第一内容标识码各自对应的第二中间信息,确定所述待监测内容是否合规的步骤,包括:
按位比对两个所述第二中间信息;
在两个所述第二中间信息完全一致的情况下,确定所述待监测内容合规;
在两个所述第二中间信息不完全一致的情况下,确定所述待监测内容不合规。
可选地,所述装置还包括:
内容属性获取模块,用于在两个所述第二中间信息不完全一致的情况下,获取两个所述第二中间信息不一致的字段,以及所述不一致的字段所标识的内容属性;
报警信号生成模块,用于基于所述不一致的字段所标识的内容属性,生成对应等级的报警信号;其中,所述内容属性包括显示设备的标识属性、播放时段属性中的至少一者。
可选地,内容标识码生成模块,包括:
编码单元,用于对所述身份编码信息迭代进行多种编码,得到每种编码下的中间信息;
第一映射单元,用于将所述多种编码各自对应的中间信息中的第一中间信息映射为预定透明图像上的多个校验点,得到所述第二内容标识码;
其中,所述第一中间信息为十进制的灰度信息,所述第一中间信息中不同位置的灰度值对应不同位置的校验点。
可选地,所述编码单元,包括;
第一转换子单元,用于将所述身份编码信息进行二进制转换;
第二转换子单元,用于将所述二进制转换得到的中间信息进行十六进制转换;
第三转换子单元,用于将十六进制转换得到的中间信息进行十进制转换,得到十进制的灰度信息。
可选地,所述预定透明图像上设置信息隐藏区,所述信息隐藏区用于携带所述身份编码信息对应的第二中间信息,所述装置还包括:
第二中间信息获取模块,用于获取所述多种编码各自对应的中间信息中的第二中间信息;所述第二中间信息早于所述第一中间信息产出,所述第二中间信息为二进制的信息;
映射模块,用于基于所述第二中间信息中每位的编码值,更新所述信息隐藏区的灰度值。
可选地,所述编码信息生成模块,包括:
内容属性获取单元,用于获取所述预定内容的多维度的内容属性,所述多维度的内容属性包括所述预定内容的名称、播放时段以及播放所述预定内容的显示设备的设备标识中的一者或多者;
编码单元,用于对所述多维度的内容属性进行三十二进制编码,得到第二预定位数的身份编码信息。
基于相同的发明构思,本公开还提供了一种计算机可读存储介质,其存储的计算机程序使得处理器执行所述的播放内容的监测方法。
基于相同的发明构思,本公开还提供了一种电子设备,包括存储器、处理器及存储在存储器上并可在处理器上运行的计算机程序,所述处理器执行时实现所述的播放内容的监测方法。
本说明书中的各个实施例均采用递进的方式描述,每个实施例重点说明的都是与其他实施例的不同之处,各个实施例之间相同相似的部分互相参见 即可。
最后,还需要说明的是,在本文中,诸如第一和第二等之类的关系术语仅仅用来将一个实体或者操作与另一个实体或操作区分开来,而不一定要求或者暗示这些实体或操作之间存在任何这种实际的关系或者顺序。而且,术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含,从而使得包括一系列要素的过程、方法、商品或者设备不仅包括那些要素,而且还包括没有明确列出的其他要素,或者是还包括为这种过程、方法、商品或者设备所固有的要素。在没有更多限制的情况下,由语句“包括一个……”限定的要素,并不排除在包括所述要素的过程、方法、商品或者设备中还存在另外的相同要素。
以上对本公开所提供的一种播放内容的监测方法、主机、***、装置、介质及设备进行了详细介绍,本文中应用了具体个例对本公开的原理及实施方式进行了阐述,以上实施例的说明只是用于帮助理解本公开的方法及其核心思想;同时,对于本领域的一般技术人员,依据本公开的思想,在具体实施方式及应用范围上均会有改变之处,综上所述,本说明书内容不应理解为对本公开的限制。
本领域技术人员在考虑说明书及实践这里公开的发明后,将容易想到本公开的其它实施方案。本公开旨在涵盖本公开的任何变型、用途或者适应性变化,这些变型、用途或者适应性变化遵循本公开的一般性原理并包括本公开未公开的本技术领域中的公知常识或惯用技术手段。说明书和实施例仅被视为示例性的,本公开的真正范围和精神由下面的权利要求指出。
应当理解的是,本公开并不局限于上面已经描述并在附图中示出的精确结构,并且可以在不脱离其范围进行各种修改和改变。本公开的范围仅由所附的权利要求来限制。
本文中所称的“一个实施例”、“实施例”或者“一个或者多个实施例”意味着,结合实施例描述的特定特征、结构或者特性包括在本公开的至少一个实施例中。此外,请注意,这里“在一个实施例中”的词语例子不一定全指同一个实施例。
在此处所提供的说明书中,说明了大量具体细节。然而,能够理解,本公开的实施例可以在没有这些具体细节的情况下被实践。在一些实例中,并 未详细示出公知的方法、结构和技术,以便不模糊对本说明书的理解。
在权利要求中,不应将位于括号之间的任何参考符号构造成对权利要求的限制。单词“包含”不排除存在未列在权利要求中的元件或步骤。位于元件之前的单词“一”或“一个”不排除存在多个这样的元件。本公开可以借助于包括有若干不同元件的硬件以及借助于适当编程的计算机来实现。在列举了若干装置的单元权利要求中,这些装置中的若干个可以是通过同一个硬件项来具体体现。单词第一、第二、以及第三等的使用不表示任何顺序。可将这些单词解释为名称。
最后应说明的是:以上实施例仅用以说明本公开的技术方案,而非对其限制;尽管参照前述实施例对本公开进行了详细的说明,本领域的普通技术人员应当理解:其依然可以对前述各实施例所记载的技术方案进行修改,或者对其中部分技术特征进行等同替换;而这些修改或者替换,并不使相应技术方案的本质脱离本公开各实施例技术方案的精神和范围。
Claims (20)
- 一种播放内容的监测方法,其特征在于,所述方法包括:在当前的监测时段,从显示设备获取显示页面截图,所述显示页面截图是对所述显示设备的显示页面进行截图得到的;从所述显示页面截图中提取第一内容标识码,所述第一内容标识码用于标识所述显示设备当前所播放的待监测内容;基于所述第一内容标识码,确定所述待监测内容是否合规。
- 根据权利要求1所述的方法,其特征在于,基于所述第一内容标识码,确定所述待监测内容是否合规,包括:基于所述第一内容标识码,进行一级校验,所述一级校验用于确定所述待监测内容是否存在异常;在所述一级校验不通过的情况下,对所述显示页面截图进行二级校验,所述二级校验用于确定所述待监测内容的异常类别;基于所述二级校验的结果,确定所述待监测内容是否合规。
- 根据权利要求2所述的方法,其特征在于,对所述显示页面截图进行二级校验,包括:对所述显示页面截图进行画面异常检测,确定所述待监测内容是否存在第一类型的异常画面和/或第二类型的异常画面;第一类型为屏幕异常的类型,第二类型为包含内容异常的类型;基于所述二级校验的结果,确定所述待监测内容是否合规,包括:若存在所述第一类型的异常画面,则确定所述待监测内容合规;若存在第二类型的异常画面,则确定所述待监测内容不合规。
- 根据权利要求3所述的方法,其特征在于,对所述显示页面截图进行画面异常检测,确定所述待监测内容是否存在第一类型的异常画面和/或第二类型的异常画面,包括:将所述显示页面截图输入至第一检测模型,并根据所述第一检测模型输出的结果确定所述待监测内容是否存在所述第一类型的异常画面;和/或,将所述显示页面截图输入至第二检测模型,并根据所述第二检测模型输出的结果确定所述待监测内容是否存在第二类型的异常画面;其中,所述第一检测模型是以多个携带有屏幕异常的图像为训练样本对第一神经网络进行训练得到的,所述第二检测模型是以多个携带异常内容的图像为训练样本对第二神经网络进行训练得到的。
- 根据权利要求2所述的方法,其特征在于,在所述一级校验不通过的情况下,对所述显示页面截图进行二级校验,包括:将所述显示页面截图发送给内容发布服务器,以使所述内容发布服务器进行所述二级校验;基于二级校验的结果,确定所述待监测内容是否合规,包括:基于所述内容发布服务器返回的二级校验的结果,确定所述待监测内容是否合规。
- 根据权利要求2-5任一所述的方法,其特征在于,在确定所述待监测内容不合规的情况下,所述方法还包括以下至少之一:关闭所述显示设备,以禁止所述显示设备进行内容播放;和/或发送报警信号,以指示所述待监测内容不合规。
- 根据权利要求1所述的方法,其特征在于,在所述在当前的监测时段,从显示设备获取显示页面截图的步骤之前,所述方法还包括:对待发送给所述显示设备进行播放的预定内容,生成与所述预定内容对应的身份编码信息;基于所述身份编码信息,生成所述预定内容对应的第二内容标识码;将所述预定内容与所述第二内容标识码发送给所述显示设备,以使所述显示设备在播放所述预定内容时,在显示页面展示所述第二内容标识码;其中,所述第二内容标识码作为所述第一内容标识码的对照,以用于确定所述待监测内容是否合规。
- 根据权利要求1所述的方法,其特征在于,基于所述第一内容标识码,确定所述待监测内容是否合规,包括:获取与所述待监测的播放时段对应的第二内容标识码,其中,所述第二内容标识码与在所述待监测的播放时段进行播放的预定内容对应;基于所述第一内容标识码和所述第二内容标识码,确定所述待监测内容是否合规。
- 根据权利要求7或8所述的方法,其特征在于,基于所述第一内容标 识码,确定所述待监测内容是否合规,包括:分别对所述第二内容标识码和所述第一内容标识码进行解析,得到所述第二内容标识码和所述第一内容标识码分别对应的第一中间信息;其中,所述第一中间信息是将内容对应的身份编码信息转换为内容标识码的过程中的中间信息;基于所述第二内容标识码和所述第一内容标识码分别对应的第一中间信息,确定所述待监测内容是否合规;其中,所述第一中间信息为所播放的内容的身份编码信息的摘要。
- 根据权利要求9所述的方法,其特征在于,所述第二内容标识码还包括隐藏区,所述隐藏区用于携带身份编码信息对应的第二中间信息,所述基于所述第二内容标识码和所述第一内容标识码分别对应的第一中间信息,确定待监测内容是否合规的步骤,包括:确定所述第二内容标识码和所述第一内容标识码分别对应的第一中间信息是否完全一致;若否,则分别对所述第二内容标识码和所述第一内容标识码的隐藏区进行解析,得到对所述第二内容标识码和所述第一内容标识码各自对应的第二中间信息;其中,所述第二中间信息在所述第一中间信息之前产出;基于所述第二内容标识码和所述第一内容标识码各自对应的第二中间信息,确定所述待监测内容是否合规。
- 根据权利要10所述的方法,其特征在于,基于所述第二内容标识码和所述第一内容标识码各自对应的第二中间信息,确定所述待监测内容是否合规,包括:按位比对两个所述第二中间信息;在两个所述第二中间信息完全一致的情况下,确定所述待监测内容合规;在两个所述第二中间信息不完全一致的情况下,确定所述待监测内容不合规。
- 根据权利要求11所述的方法,其特征在于,在两个所述第二中间信息不完全一致的情况下,所述方法还包括:获取两个所述第二中间信息不一致的字段,以及所述不一致的字段所标识的内容属性;基于所述不一致的字段所标识的内容属性,生成对应等级的报警信号;其中,所述内容属性包括显示设备的标识属性、播放时段属性中的至少一者。
- 根据权利要求7所述的方法,其特征在于,基于所述身份编码信息,生成所述预定内容的第二内容标识码,包括:对所述身份编码信息迭代进行多种编码,得到每种编码下的中间信息;将所述多种编码各自对应的中间信息中的第一中间信息映射为预定透明图像上的多个校验点,得到所述第二内容标识码;其中,所述第一中间信息为十进制的灰度信息,所述第一中间信息中不同位置的数值对应不同位置的校验点。
- 根据权利要求13所述的方法,其特征在于,所述对所述身份编码信息迭代进行多种编码,得到每种编码下的中间信息的步骤,包括;将所述身份编码信息进行二进制转换;将所述二进制转换得到的中间信息进行十六进制转换;将十六进制转换得到的中间信息进行十进制转换,得到十进制的灰度信息。
- 根据权利要求13或14所述的方法,其特征在于,所述预定透明图像上设置信息隐藏区,所述信息隐藏区用于携带所述身份编码信息对应的第二中间信息,所述方法还包括:获取所述多种编码各自对应的中间信息中的第二中间信息;所述第二中间信息早于所述第一中间信息产出,所述第二中间信息为二进制的信息;基于所述第二中间信息中每位的编码值,更新所述信息隐藏区的灰度值。
- 一种播控主机,其特征在于,所述播控主机用于执行权利要求1-16任一所述的播放内容的监测方法。
- 一种监播***,其特征在于,包括:内容发布服务器,与所述内容发布服务器连接的至少一个播控主机,以及与每个所述播控主机连接的显示设备;所述内容发布服务器,用于将待播放的内容发送给对应的播控主机;所述播控主机,用于将所述待播放的内容发送给所述显示设备,并用于执行权利要求1-15任一所述的播放内容的监测方法;所述显示设备,用于播放所述待播放的内容。
- 一种播放内容的监测装置,其特征在于,所述装置包括:截图获取模块,用于在当前的监测时段,从显示设备获取显示页面截图,所述显示页面截图是对所述显示设备当前播放的待监测内容进行截图得到的;标识码获取模块,用于从所述显示页面截图中提取第一内容标识码,所述第一内容标识码用于标识所播放的内容;验证模块,用于基于所述第一内容标识码,确定所述待监测内容是否合规。
- 一种计算机可读存储介质,其特征在于,其存储的计算机程序使得处理器执行如权利要求1-15任一所述的播放内容的监测方法。
- 一种电子设备,其特征在于,包括存储器、处理器及存储在存储器上并可在处理器上运行的计算机程序,所述处理器执行时实现如权利要求1-15任一所述的播放内容的监测方法。
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/CN2022/116276 WO2024045064A1 (zh) | 2022-08-31 | 2022-08-31 | 播放内容的监测方法、主机、***、装置、介质及设备 |
CN202280002992.4A CN117957847A (zh) | 2022-08-31 | 2022-08-31 | 播放内容的监测方法、主机、***、装置、介质及设备 |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/CN2022/116276 WO2024045064A1 (zh) | 2022-08-31 | 2022-08-31 | 播放内容的监测方法、主机、***、装置、介质及设备 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2024045064A1 true WO2024045064A1 (zh) | 2024-03-07 |
Family
ID=90099971
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2022/116276 WO2024045064A1 (zh) | 2022-08-31 | 2022-08-31 | 播放内容的监测方法、主机、***、装置、介质及设备 |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN117957847A (zh) |
WO (1) | WO2024045064A1 (zh) |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109885446A (zh) * | 2018-12-25 | 2019-06-14 | 北京互金新融科技有限公司 | 确定网站页面显示状态的方法及装置 |
CN112148896A (zh) * | 2020-09-10 | 2020-12-29 | 京东数字科技控股股份有限公司 | 一种用于终端媒体监播的数据处理方法及装置 |
CN114466210A (zh) * | 2022-01-27 | 2022-05-10 | 广州方硅信息技术有限公司 | 直播质量检测处理方法及其装置、设备、介质、产品 |
WO2022127335A1 (zh) * | 2020-12-17 | 2022-06-23 | 中兴通讯股份有限公司 | 信息监测方法、终端、服务器和计算机可读存储介质 |
CN114881708A (zh) * | 2022-05-31 | 2022-08-09 | 京东方科技集团股份有限公司 | 监测投放内容的方法、装置、***、设备及存储介质 |
-
2022
- 2022-08-31 CN CN202280002992.4A patent/CN117957847A/zh active Pending
- 2022-08-31 WO PCT/CN2022/116276 patent/WO2024045064A1/zh active Application Filing
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109885446A (zh) * | 2018-12-25 | 2019-06-14 | 北京互金新融科技有限公司 | 确定网站页面显示状态的方法及装置 |
CN112148896A (zh) * | 2020-09-10 | 2020-12-29 | 京东数字科技控股股份有限公司 | 一种用于终端媒体监播的数据处理方法及装置 |
WO2022127335A1 (zh) * | 2020-12-17 | 2022-06-23 | 中兴通讯股份有限公司 | 信息监测方法、终端、服务器和计算机可读存储介质 |
CN114466210A (zh) * | 2022-01-27 | 2022-05-10 | 广州方硅信息技术有限公司 | 直播质量检测处理方法及其装置、设备、介质、产品 |
CN114881708A (zh) * | 2022-05-31 | 2022-08-09 | 京东方科技集团股份有限公司 | 监测投放内容的方法、装置、***、设备及存储介质 |
Also Published As
Publication number | Publication date |
---|---|
CN117957847A (zh) | 2024-04-30 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110837615A (zh) | 广告内容信息过滤人工智能审核*** | |
KR20130010472A (ko) | 마스크 기반 시도 응답 테스트 | |
Lago et al. | Visual and textual analysis for image trustworthiness assessment within online news | |
CN110502512B (zh) | 信息载体生成方法、关联方法、装置、设备及存储介质 | |
CN112163072A (zh) | 基于多数据源的数据处理方法以及装置 | |
CN114840286B (zh) | 基于大数据的业务处理方法及服务器 | |
CN107769958A (zh) | 基于日志的服务器网络安全事件自动化分析方法及*** | |
CN106815779A (zh) | 建设工程质量检测防假样***及方法 | |
CN115982012A (zh) | 一种接口管理能力成熟度的评估模型及方法 | |
CN114282882B (zh) | 素材工单审核方法、装置、设备及存储介质 | |
Qu | Research on password detection technology of iot equipment based on wide area network | |
WO2024045064A1 (zh) | 播放内容的监测方法、主机、***、装置、介质及设备 | |
CN117195319A (zh) | 保函文件电子件的验真方法、装置、电子设备和介质 | |
CN112347079A (zh) | 一种数据库内容保密检查***及检查方法 | |
CN116402334A (zh) | 一种多模态数据合规分析与智能评估方法及装置 | |
CN114265759A (zh) | 一种数据信息泄露后的溯源方法、***及电子设备 | |
CN112291625B (zh) | 信息质量处理方法、装置、电子设备及存储介质 | |
CN115473669A (zh) | 一种用于区块链的可视化展示方法 | |
CN111340038B (zh) | 用于mock测试的一次性影像数据获取方法 | |
CN113177617A (zh) | 一种防篡改能效识别码的识别方法及*** | |
WO2023220956A1 (zh) | 多媒体监播方法、***及电子设备 | |
CN206921147U (zh) | 建设工程质量检测防假样*** | |
CN117389769B (zh) | 基于云服务的浏览器端富文本拷贝方法、***及云平台 | |
Sun et al. | [Retracted] Construction of Live Broadcast Training Platform Based on “Cloud Computing” and “Big Data” and “Wireless Communication Technology” | |
CN113014591B (zh) | 假冒公众号的检测方法和装置、电子设备、及介质 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
WWE | Wipo information: entry into national phase |
Ref document number: 202280002992.4 Country of ref document: CN |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 22956892 Country of ref document: EP Kind code of ref document: A1 |