CN111277877A - Multimedia display large-screen safety protection system and method based on content identification - Google Patents
Multimedia display large-screen safety protection system and method based on content identification Download PDFInfo
- Publication number
- CN111277877A CN111277877A CN201811381586.5A CN201811381586A CN111277877A CN 111277877 A CN111277877 A CN 111277877A CN 201811381586 A CN201811381586 A CN 201811381586A CN 111277877 A CN111277877 A CN 111277877A
- Authority
- CN
- China
- Prior art keywords
- module
- video
- video stream
- riot
- pornographic
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 20
- 238000000605 extraction Methods 0.000 claims abstract description 19
- 230000005540 biological transmission Effects 0.000 claims abstract description 13
- 238000012550 audit Methods 0.000 claims abstract 2
- 230000000903 blocking effect Effects 0.000 claims description 11
- 239000000284 extract Substances 0.000 claims description 2
- 230000003139 buffering effect Effects 0.000 claims 1
- 238000004880 explosion Methods 0.000 claims 1
- 238000013527 convolutional neural network Methods 0.000 description 7
- 238000012015 optical character recognition Methods 0.000 description 4
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000007599 discharging Methods 0.000 description 2
- 238000010276 construction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 239000000428 dust Substances 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000010304 firing Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000000306 recurrent effect Effects 0.000 description 1
- 230000006403 short-term memory Effects 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Images
Classifications
-
- 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
- H04N21/41—Structure of client; Structure of client peripherals
- H04N21/414—Specialised client platforms, e.g. receiver in car or embedded in a mobile appliance
- H04N21/41415—Specialised client platforms, e.g. receiver in car or embedded in a mobile appliance involving a public display, viewable by several users in a public space outside their home, e.g. movie theatre, information kiosk
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/41—Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items
-
- 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
- H04N21/44—Processing of video elementary streams, e.g. splicing a video clip retrieved from local storage with an incoming video stream or rendering scenes according to encoded video stream scene graphs
- H04N21/44008—Processing of video elementary streams, e.g. splicing a video clip retrieved from local storage with an incoming video stream or rendering scenes according to encoded video stream scene graphs involving operations for analysing video streams, e.g. detecting features or characteristics in the video stream
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N7/00—Television systems
- H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Software Systems (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- Data Mining & Analysis (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Biophysics (AREA)
- Mathematical Physics (AREA)
- Biomedical Technology (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Information Transfer Between Computers (AREA)
Abstract
The invention provides a multimedia display large-screen safety protection system and method based on content identification. The system comprises an input module, an output module, a buffer pool, a frame extraction strategy module, a pornography identification module, a sensitive character identification module, an riot and terrorist identification module and a decision module. The input module receives an input, a standard video stream, from a video transmission port. And the output module receives the video stream from the buffer pool and converts the video stream into a video signal or a video stream message of each video transmission port. And the buffer pool cuts the video stream and establishes an FIFO queue. And the frame extraction strategy module converts the video stream into a picture set according to the frame extraction strategy. The pornographic recognition module recognizes pornographic degrees. The sensitive character recognition module recognizes the sensitive characters. The riot and terrorist identification module identifies the riot and terrorist information. And the decision module calculates an audit result according to the requirement. The system and the method adopt a buffer mechanism, thereby reducing the influence on the video fluency; the method is arranged close to a large multimedia display screen, inserted into the video stream of the original system and takes over the passing of the final video signal, so that the pornographic information, sensitive characters and violence information brought by misoperation, malicious operation and malicious attack are limited finally from the destination end.
Description
Technical Field
The invention relates to a multimedia display large-screen safety protection system and method based on content identification.
Abbreviations and noun explanations:
CNN: called the Convolutional Neural Network, overall.
RNN: called current Neural Networks, Recurrent Neural Networks.
ResNet 50: a residual network design with a depth of 50 layers.
Darknet 19: a convolutional neural network design with 19 convolutional kernels and 5 max-firing.
OCR: optical Character Recognition, the full name of Optical Character Recognition. Refers to a process in which an electronic device (e.g., a scanner or a digital camera) checks a character printed on paper, determines its shape by detecting dark and light patterns, and then translates the shape into a computer text by a character recognition method; the method is characterized in that characters in a paper document are converted into an image file with a black-white dot matrix in an optical mode aiming at print characters, and the characters in the image are converted into a text format through recognition software for further editing and processing by word processing software.
LSTM: the Long Short Term Memory networks are called. The method is a special RNN structure and can learn long-term dependence.
Background
The multimedia display large screen is mainly applied to the fields of military command, electric power, traffic monitoring, studios and the like at present, and further enters the markets of security protection, commercial exhibition and the like along with the development of smart cities and the like. The large multimedia display screen is used as a large terminal display platform of a smart city, plays a role in many aspects, and particularly has the advantages of cold resistance, water resistance, dust resistance, high brightness, splicing property, easiness in maintenance and the like in the field of smart traffic so that the large multimedia display screen can stand out in the field of traffic display terminals and becomes an inevitable choice for the construction of the large smart city traffic screen. A large number of multimedia display screens are arranged in a public space and used for displaying specific information, and the multimedia display screens can be used as attack targets to play pornographic information, sensitive characters and violence and terrorism information for the purpose of public opinion attack, so that negative social influence is caused.
The existing multimedia display large-screen playing system has simple playing mechanism and the following serious defects for the insufficient identification of the playing content:
1. there is no recognition mechanism for the playing content. When the multimedia display large-screen playing system plays the set content, the playing content is not identified, and the risk of misoperation or malicious playing of pornographic information, sensitive characters and violence and terrorism information exists.
2. There is no authentication mechanism for the broadcast content distribution terminal. An attacker can imitate the malicious playing of pornographic information, sensitive characters and riot and terrorist information by the publishing terminal in the system, and the legality of the playing content source cannot be effectively identified by the large multimedia display screen, so that the risk of mistaken playing exists.
Disclosure of Invention
The invention provides a multimedia display large-screen safety protection system and method based on content identification, which can realize identification of pornographic information, sensitive characters and violence and terrorism information in display content and block display by using various content identification technologies based on the buffer of the display content of a multimedia display large-screen so as to avoid negative public influence caused by broadcasting the information through the multimedia display large-screen.
The multimedia display large-screen safety protection system and method based on content identification can protect the content safety of a multimedia display large screen from three layers.
1. And identifying the pornographic information of the played content. And information such as pornographic lenses, pornographic pictures and the like in the played content is prevented.
2. And performing sensitive character recognition on the played content. And sensitive characters such as secret-related characters and anti-motion characters are prevented from appearing in the played content.
3. And identifying the riot information of the played content. And information such as violence, terrorism and the like in the played content is prevented.
Fig. 1 is a system and method for multimedia display large screen security protection based on content identification according to the present invention, which includes an input module, an output module, a buffer pool, a frame-extracting strategy module, a pornography identification module, a sensitive character identification module, an riot identification module, and a decision module. The system adopts a buffer mechanism, so that the influence on the video fluency is reduced; the method is arranged close to a large multimedia display screen, inserted into the video stream of the original system and takes over the passing of the final video signal, so that the pornographic information, sensitive characters and violence information brought by misoperation, malicious operation and malicious attack are limited finally from the destination end.
The input module receives input from a video transmission port, including VGA, DVI and HDMI video signals, custom video signals transmitted on an RJ45 port, video stream messages transmitted by a video transmission protocol based on IP on an RJ45 port, converts the video signals or the video stream messages into H.264/H.265 standard video, outputs the video signals or the video stream messages to a buffer pool, and simultaneously performs flow control according to the occupation state of the buffer pool.
The output module receives the H.264/H.265 standard video from the buffer pool, converts the video into video signals or video stream messages of each video transmission port, wherein the video signals comprise VGA, DVI and HDMI video signals, custom video signals transmitted on RJ45 ports, and video stream messages transmitted by a video transmission protocol based on IP on RJ45 ports are sent to a rear-end multimedia display large screen for display.
The buffer pool receives the video stream sent by the input module, cuts the video stream into blocks according to the set identification time length, and arranges the blocks into an input FIFO queue. And taking out a video stream block from the front part of the FIFO queue to be checked, sending the video stream block into the frame-extracting strategy module for imaging, entering a region to be judged to receive a passing/blocking instruction from the decision module, discharging the passing video stream block into the output FIFO queue, and discharging the blocked video stream block into the FIFO queue instead of the blocking prompt video. And the output FIFO queue sequentially outputs the video stream to the output module.
And the frame extraction strategy module is used for receiving the video stream blocks sent by the buffer pool and converting the video stream blocks into a picture set according to the frame extraction strategy. There are three frame extraction strategies: 1. strictly extracting each frame of picture; 2. extracting only I-frame pictures containing main information; 3. and randomly extracting picture frames according to the parameter setting. Which frame extraction strategy is chosen depends on the identification requirements and system resources. And the converted picture set is respectively sent to a pornographic recognition module, a sensitive character recognition module and an riot and terrorist recognition module for recognition.
And the pornographic identification module receives the picture set sent by the frame extraction strategy module and sends the picture set into a trained ResNet50 neural network model. ResNet50 uses a connection mode called "short connection", abstracts the characteristics of the input image through 50 layers of convolution layers, calculates the pornographic indexes of the picture through model parameters in the full connection layer by the high-order characteristics, and sends the indexes into the decision module, wherein the index range is (0, 1).
The sensitive character recognition module receives the picture set sent by the frame extraction strategy module, and the picture set is recognized into single characters/words through the OCR unit, and the four steps are carried out: 1. analyzing the connected region, and detecting the region outline and the sub-outline of the character region to form a block region; 2. finding a block area, detecting a character outline to obtain a text line, and obtaining single characters/words through blanks; 3. finding out text lines and single characters/words, adopting a self-adaptive classifier to analyze the single characters/words, and carrying out single character/word analysis twice; 4. and obtaining a text, and identifying a blank containing fuzzy, stroke height and lower case letters. And matching and comparing the sensitive character library to judge whether sensitive characters exist or not, wherein the result is 0 or 1, and the result is sent to a decision module.
The violence and terrorist identification module receives the picture set sent by the frame extraction strategy module and sends the picture set into the trained Darknet19+ CNN + LSTM model. The two selected input frames are processed by two darknets 19 trained in advance, the two frame outputs from the bottom layer of the Darknet19 pre-trained model are concatenated in the last pass and then fed into an additional CNN, which should learn local motion features as well as appearance invariant features by comparing the two frame feature maps since the outputs from the bottom layer are considered as low-level features. The two frame outputs from the top layer of the Darknet19 pre-trained network are also connected and fed to another additional CNN to compare the advanced features of the two frames. The outputs from the two additional CNNs are then concatenated and passed to the fully-connected layer and LSTM unit to learn the global temporal features. Finally, the output of the LSTM unit is classified by a fully connected layer containing two neurons representing two classes (riot and non-riot), respectively, giving a prediction of whether or not to contain riot information, the result being in the (0, 1) interval, while the result is fed into the decision module.
The decision module outputs a blocking command to a to-be-determined area of the buffer pool according to results sent by the pornographic recognition module, the sensitive character recognition module and the riot and terrible recognition module and by combining pornographic requirements of users on pornographic degree thresholds and riot and terrible degree thresholds, when the result of the sensitive character is greater than the pornographic degree threshold and the result of the sensitive character is greater than the riot and terrible degree threshold is 1; and when the (pornography > pornography threshold) | | sensitive character result | | | (riot terrorism > riot terrorism threshold) | 0, outputting a passing command to a region to be judged of the buffer pool.
The invention relates to a multimedia display large-screen safety protection method based on content identification, which is characterized in that an input module takes over a video signal transmitted by a multimedia display large-screen system or a video stream message transmitted by a video transmission protocol based on IP (Internet protocol) and outputs an H.264/H.265 standard video to a buffer pool by adopting the multimedia display large-screen safety protection system based on content identification; the buffer pool cuts the standard video into blocks, sends the blocks into an area to be judged and simultaneously sends the blocks into a frame extraction strategy module; the frame extraction strategy module extracts video stream blocks into a picture set and respectively sends the picture set into a pornographic recognition module, a sensitive character recognition module and a riot and terrorist recognition module; the pornographic recognition module calculates the pornographic degree of the picture set, outputs an index of a (0, 1) interval, and sends the index into the decision module when the larger the pornographic degree is, the higher the pornographic degree is; the sensitive character recognition module recognizes and compares characters in the picture set, outputs 0 or 1, 1 representing the existence of sensitive characters, and outputs the sensitive characters to the decision module; the riot and terrorist identification module calculates the riot and terrorist degree of the picture set, and outputs indexes of (0, 1) intervals, wherein the larger the index is, the higher the riot and terrorist degree is, and the larger the index is, the higher the riot and terrorist degree is, the larger the index is, the higher the; the decision module calculates a final blocking/passing result according to the sent results of the three identification modules, and inputs the final blocking/passing result to a module to be judged of the buffer pool; the buffer pool to-be-determined module determines to output the alternative video/the video to the output queue according to the sent blocking/passing instruction; the output module takes out the standard video from the output queue of the buffer pool, converts the standard video into a video signal or a video stream message which is based on IP and transmitted by a video transmission protocol, and sends the video stream message back to the next stage of the multimedia display large-screen system through an interface.
The specific implementation mode is as follows:
fig. 2 is a deployment of the multimedia display large-screen security system and method based on content identification according to the present invention. The system and the method are arranged between the sending equipment and the receiving equipment in a mode of being closest to a large multimedia display screen, take over the passing of the final video signal, and finally limit the pornographic information, sensitive characters and violence information brought by misoperation, malicious operation and malicious attack from the destination end.
Drawings
Fig. 1 is a schematic diagram of a multimedia display large-screen security protection system and method based on content identification according to the present invention.
Fig. 2 is a schematic diagram of a deployment mode of the multimedia display large-screen security protection system and method based on content identification according to the present invention.
Claims (9)
1. A multimedia display large screen safety protection system based on content identification is characterized in that the system is arranged between a sending device and a receiving device in a mode closest to a multimedia display large screen, and takes over the passing of a final video signal, so that pornographic information, sensitive characters and violence information brought by misoperation, malicious operation and malicious attack are limited finally from a destination end; the video streaming buffer mechanism is adopted and comprises an input module, an output module, a buffer pool, a frame-extracting strategy module, a pornography identification module, a sensitive character identification module, an explosion and terrorism identification module and a decision module, wherein:
A. the input module converts the video signal or the video stream message into an H.264/H.265 standard video;
B. the output module converts the H.264/H.265 standard video into video signals or video stream messages of each video transmission port, and sends the video signals or video stream messages to a rear-end multimedia display large screen for display;
C. the buffer pool completes the buffer input and buffer output functions of the video stream;
D. the frame extraction strategy module is used for converting the frame extraction strategy into a picture set according to the frame extraction strategy;
E. the pornographic recognition module gives out pornographic indexes;
F. the sensitive character recognition module gives out whether the sensitive characters are contained or not;
G. the riot and terrorist identification module gives out an riot and terrorist degree index;
H. and the decision module calculates whether the video streaming block passes the audit or not according to the results sent by the pornography identification module, the sensitive character identification module and the riot and terrorist identification module and the requirements of the user on the pornography threshold value and the riot and terrorist threshold value.
2. The system of claim 1, wherein the system employs a video stream buffering mechanism to reduce the impact on video smoothness.
3. The system of claim 1, wherein the system is deployed in close proximity to the large multimedia display screen, inserted into the video stream of the original system, and takes over the passage of the final video signal.
4. The system of claim 1, wherein the input module identifies video signals or video stream messages from various video ports, including VGA, DVI, HDMI video signals, custom video signals transmitted through RJ45 port, video stream messages transmitted through RJ45 port based on IP and video transmission protocol, and converts the video stream messages into h.264/h.265 standard video.
5. The system of claim 1, wherein the output module converts the determined h.264/h.265 standard video into video signals or video stream messages output from each video transmission port, and the output module comprises VGA, DVI, HDMI, and RJ45 ports, and the RJ45 port comprises IP video stream messages and video signals transmitted through the RJ45 port.
6. The system as claimed in claim 1, wherein the buffer pool divides the video stream into blocks, and sets an input FIFO queue, an output FIFO queue, and a to-be-determined area, the video stream blocks of the input FIFO queue are sequentially sent to a next module for determination and simultaneously enter the to-be-determined area, the to-be-determined area receives a pass/block command from the decision module, the passing video stream blocks are arranged in the output FIFO queue, and the blocked video stream blocks are arranged in the FIFO queue instead of the blocking prompt video. And the output FIFO queue sequentially outputs the video stream to the output module.
7. The system of claim 1, wherein the frame extraction strategy module selects the frame extraction strategy according to the identification requirement and system resources, and comprises: strictly extracting each frame of picture; or only extracting the I frame picture containing the main information; or randomly extracting the picture frame according to the parameter setting.
8. The multimedia display large-screen security protection system based on content identification as claimed in claim 1, wherein the decision module calculates whether to allow the video stream block to be displayed on the multimedia display large screen in combination with the pornographic requirements of the user on the pornographic threshold and the riot degree threshold, and outputs a blocking command to the to-be-determined area of the buffer pool when (pornography > pornographic threshold) | | sensitive text result | | | (riot degree > riot degree threshold) | 1; and when the (pornography > pornography threshold) | | sensitive character result | | | (riot terrorism > riot terrorism threshold) | 0, outputting a passing command to a region to be judged of the buffer pool.
9. The invention relates to a multimedia display large-screen security protection method based on content identification, which is characterized in that the invention adopts a multimedia display large-screen security protection system based on content identification as claimed in any one of claims 1 to 8, an input module takes over a video signal transmitted by the multimedia display large-screen system or a video stream message transmitted by a video transmission protocol based on IP, and outputs an H.264/H.265 standard video to a buffer pool; the buffer pool cuts the standard video into blocks, sends the blocks into an area to be judged and simultaneously sends the blocks into a frame extraction strategy module; the frame extraction strategy module extracts video stream blocks into a picture set and respectively sends the picture set into a pornographic recognition module, a sensitive character recognition module and a riot and terrorist recognition module; the pornographic recognition module calculates the pornographic degree of the picture set, outputs an index of a (0, 1) interval, and sends the index into the decision module when the larger the pornographic degree is, the higher the pornographic degree is; the sensitive character recognition module recognizes and compares characters in the picture set, outputs 0 or 1, 1 representing the existence of sensitive characters, and outputs the sensitive characters to the decision module; the riot and terrorist identification module calculates the riot and terrorist degree of the picture set, and outputs indexes of (0, 1) intervals, wherein the larger the index is, the higher the riot and terrorist degree is, and the larger the index is, the higher the riot and terrorist degree is, the larger the index is, the higher the; the decision module calculates a final blocking/passing result according to the sent results of the three identification modules, and inputs the final blocking/passing result to a module to be judged of the buffer pool; the buffer pool to-be-determined module determines to output the alternative video/the video to the output queue according to the sent blocking/passing instruction; the output module takes out the standard video from the output queue of the buffer pool, converts the standard video into a video signal or a video stream message which is based on IP and transmitted by a video transmission protocol, and sends the video stream message back to the next stage of the multimedia display large-screen system through an interface.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811381586.5A CN111277877A (en) | 2018-11-20 | 2018-11-20 | Multimedia display large-screen safety protection system and method based on content identification |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811381586.5A CN111277877A (en) | 2018-11-20 | 2018-11-20 | Multimedia display large-screen safety protection system and method based on content identification |
Publications (1)
Publication Number | Publication Date |
---|---|
CN111277877A true CN111277877A (en) | 2020-06-12 |
Family
ID=70998479
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811381586.5A Pending CN111277877A (en) | 2018-11-20 | 2018-11-20 | Multimedia display large-screen safety protection system and method based on content identification |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111277877A (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111967493A (en) * | 2020-06-29 | 2020-11-20 | 北京百度网讯科技有限公司 | Image auditing method and device, electronic equipment and storage medium |
CN112040319A (en) * | 2020-09-03 | 2020-12-04 | 上海物盾信息科技有限公司 | Intelligent yellow identification method for advertisement screen and firewall system |
CN112380364A (en) * | 2020-11-17 | 2021-02-19 | 平安养老保险股份有限公司 | Method and system for file authentication |
CN113660291A (en) * | 2021-10-18 | 2021-11-16 | 杭州海康威视数字技术股份有限公司 | Method and device for preventing malicious tampering of intelligent large-screen display information |
CN115834939A (en) * | 2023-02-21 | 2023-03-21 | 北京匠数科技有限公司 | Electronic screen information monitoring equipment and method and electronic screen content safety pre-examination system |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105979359A (en) * | 2016-06-24 | 2016-09-28 | 中国人民解放军63888部队 | Video output control method and device based on content detection |
CN107566903A (en) * | 2017-09-11 | 2018-01-09 | 北京匠数科技有限公司 | A kind of video filtering device and method, video display system |
-
2018
- 2018-11-20 CN CN201811381586.5A patent/CN111277877A/en active Pending
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105979359A (en) * | 2016-06-24 | 2016-09-28 | 中国人民解放军63888部队 | Video output control method and device based on content detection |
CN107566903A (en) * | 2017-09-11 | 2018-01-09 | 北京匠数科技有限公司 | A kind of video filtering device and method, video display system |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111967493A (en) * | 2020-06-29 | 2020-11-20 | 北京百度网讯科技有限公司 | Image auditing method and device, electronic equipment and storage medium |
CN112040319A (en) * | 2020-09-03 | 2020-12-04 | 上海物盾信息科技有限公司 | Intelligent yellow identification method for advertisement screen and firewall system |
CN112380364A (en) * | 2020-11-17 | 2021-02-19 | 平安养老保险股份有限公司 | Method and system for file authentication |
CN113660291A (en) * | 2021-10-18 | 2021-11-16 | 杭州海康威视数字技术股份有限公司 | Method and device for preventing malicious tampering of intelligent large-screen display information |
CN115834939A (en) * | 2023-02-21 | 2023-03-21 | 北京匠数科技有限公司 | Electronic screen information monitoring equipment and method and electronic screen content safety pre-examination system |
CN115834939B (en) * | 2023-02-21 | 2023-05-12 | 北京匠数科技有限公司 | Electronic screen information monitoring equipment and method and electronic screen content security pre-examination system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111277877A (en) | Multimedia display large-screen safety protection system and method based on content identification | |
US20230259553A1 (en) | Scene annotation and action description using machine learning | |
KR102683700B1 (en) | Video processing method, apparatus, electronic device and storage medium and computer program | |
US8073203B2 (en) | Generating effects in a webcam application | |
Yang et al. | Sf-net: Structured feature network for continuous sign language recognition | |
US10304458B1 (en) | Systems and methods for transcribing videos using speaker identification | |
US10977872B2 (en) | Graphical style modification for video games using machine learning | |
CN109218629B (en) | Video generation method, storage medium and device | |
US20190362171A1 (en) | Living body detection method, electronic device and computer readable medium | |
CN110796098B (en) | Method, device, equipment and storage medium for training and auditing content auditing model | |
CN113424554A (en) | Text annotation of sound effects | |
US11334973B2 (en) | Image colorizing method and device | |
CN115546489B (en) | Multi-mode image semantic segmentation method based on cross-mode feature reinforcement and interaction | |
CN109035147A (en) | Image processing method and device, electronic device, storage medium and computer equipment | |
US20230199280A1 (en) | Machine learning device and image processing device | |
CN115238105A (en) | Illegal content detection method, system, equipment and medium fusing multimedia | |
CN102629969A (en) | Smear eliminating method during shooting of plane objects | |
WO2017116015A1 (en) | Content recognition technology-based automatic content generation method and system | |
CN115393765A (en) | Video subtitle translation method, device and equipment | |
CN109977891A (en) | A kind of object detection and recognition method neural network based | |
CN115474088A (en) | Video processing method, computer equipment and storage medium | |
US20230084980A1 (en) | System for detecting face liveliness in an image | |
TWI698757B (en) | Smart engine with dynamic profiles and method of operating smart engine | |
Animesh et al. | Fast text detection from single hazy image using smart device | |
CN109714569B (en) | Real-time fusion method and device for multi-channel video images |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200612 |