CN113643136B - Information processing method, system and device - Google Patents
Information processing method, system and device Download PDFInfo
- Publication number
- CN113643136B CN113643136B CN202111020380.1A CN202111020380A CN113643136B CN 113643136 B CN113643136 B CN 113643136B CN 202111020380 A CN202111020380 A CN 202111020380A CN 113643136 B CN113643136 B CN 113643136B
- Authority
- CN
- China
- Prior art keywords
- picture
- picture data
- prediction model
- scene type
- management platform
- 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.)
- Active
Links
- 230000010365 information processing Effects 0.000 title claims abstract description 40
- 238000003672 processing method Methods 0.000 title claims abstract description 23
- 238000002372 labelling Methods 0.000 claims abstract description 54
- 238000012549 training Methods 0.000 claims abstract description 50
- 238000000034 method Methods 0.000 claims abstract description 35
- 238000010586 diagram Methods 0.000 claims description 24
- 238000004891 communication Methods 0.000 claims description 13
- 238000004590 computer program Methods 0.000 claims description 10
- 230000004044 response Effects 0.000 claims description 8
- 238000012545 processing Methods 0.000 abstract description 7
- 238000010801 machine learning Methods 0.000 abstract description 6
- 230000008569 process Effects 0.000 description 6
- 229910000831 Steel Inorganic materials 0.000 description 5
- 230000006870 function Effects 0.000 description 5
- 239000010959 steel Substances 0.000 description 5
- 230000000694 effects Effects 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 241000282887 Suidae Species 0.000 description 3
- 230000004927 fusion Effects 0.000 description 3
- 239000002184 metal Substances 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 238000005070 sampling Methods 0.000 description 3
- 239000010779 crude oil Substances 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 239000013307 optical fiber Substances 0.000 description 2
- 230000000644 propagated effect Effects 0.000 description 2
- 241000283690 Bos taurus Species 0.000 description 1
- 101100400452 Caenorhabditis elegans map-2 gene Proteins 0.000 description 1
- 229910001294 Reinforcing steel Inorganic materials 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000012790 confirmation Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- 238000007499 fusion processing Methods 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
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
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/04—Trading; Exchange, e.g. stocks, commodities, derivatives or currency exchange
-
- 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
- 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
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
- G06Q30/0202—Market predictions or forecasting for commercial activities
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Finance (AREA)
- Accounting & Taxation (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Development Economics (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Biomedical Technology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- Artificial Intelligence (AREA)
- Entrepreneurship & Innovation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- General Business, Economics & Management (AREA)
- Economics (AREA)
- Marketing (AREA)
- Game Theory and Decision Science (AREA)
- Technology Law (AREA)
- Image Analysis (AREA)
Abstract
The embodiment of the disclosure discloses an information processing method, an information processing system and an information processing device. One embodiment of the method comprises the following steps: receiving picture data and scene types corresponding to the picture data sent by a picture management platform; responding to the absence of a picture prediction model corresponding to the scene type, when judging that the picture data meets the data requirement of model training, taking the picture data as input, taking a labeling result graph labeled as a target area in the picture data as output, and training to obtain the picture prediction model corresponding to the scene type; and storing the picture prediction model and the corresponding relation between the picture prediction model and the scene type. A method, a system and a device for processing information aiming at multiple scenes based on machine learning are realized.
Description
Technical Field
Embodiments of the present disclosure relate to the field of computer technology, and in particular, to the field of data processing technology, and in particular, to an information processing method, system, and apparatus.
Background
China is the global largest commodity futures market and first large agricultural futures market. The number of varieties on the market is close to 80, and the market comprises agricultural and sideline products such as live pigs, live cattle and the like, metal products such as steel bars, steel pipes and the like, and energy products such as crude oil, rubber and the like, so that the stability of the futures trade market can directly influence the national economic system. In order to ensure the normal running of a large number of commodity transactions, in the commodity futures transactions, the actual storage condition of the objects needs to be timely and accurately determined, and an inventory system aiming at different objects needs to be realized.
Disclosure of Invention
The embodiment of the disclosure provides an information processing method, an information processing system and an information processing device.
In a first aspect, an embodiment of the present disclosure provides an information processing method, including: receiving picture data and scene types corresponding to the picture data sent by a picture management platform; responding to the absence of a picture prediction model corresponding to the scene type, when judging that the picture data meets the data requirement of model training, taking the picture data as input, taking a labeling result graph labeled as a target area in the picture data as output, and training to obtain the picture prediction model corresponding to the scene type; and storing the picture prediction model and the corresponding relation between the picture prediction model and the scene type.
In some embodiments, the method further comprises: and in response to the existence of the picture prediction model corresponding to the scene type, inputting the picture data into the picture prediction model, and generating a labeling result diagram corresponding to the input picture data and the number of target areas in the labeling result diagram.
In some embodiments, the method further comprises: and sending the picture data, the labeling result graph corresponding to the picture data and the number of the target areas corresponding to the labeling result to a picture management platform.
In some embodiments, taking the picture data as input and the labeling result graph labeled as the target area in the picture data as output, training to obtain a picture prediction model corresponding to the scene type includes: and taking the picture data as input and the blob map marked as the target communication area in the picture data as output, and training to obtain a picture prediction model corresponding to the scene type.
In some embodiments, the network structure of the picture prediction model is constructed based on a depth residual network and a feature map pyramid network; the depth residual network is resnet-50 and the feature map pyramid network is a recursive FPN.
In some embodiments, the method further comprises: and sending prompt information corresponding to the model training result to the picture management platform.
In a second aspect, embodiments of the present disclosure provide an information processing system including: and the model management platform is used for executing the information processing method of any one of the above.
In some embodiments, the system further comprises: a picture management platform; the picture management platform is used for receiving the video stream sent by the picture acquisition device and acquiring picture data in the video stream; determining a scene type corresponding to the picture data according to the configuration information of the picture acquisition device; and sending the picture data and the scene type to a model management platform.
In some embodiments, the image management platform is further configured to receive image data sent by the model management platform, a labeling result graph corresponding to the image data, and a number of target areas corresponding to the labeling result; and sending the received picture data, the labeling result diagram and the number of the target areas to the terminal.
In some embodiments, the image management platform is further configured to perform scene type configuration on each accessed image acquisition device; and managing the scene types of the image acquisition devices.
In a third aspect, an embodiment of the present disclosure provides an information processing apparatus including: the receiving unit is configured to receive the picture data sent by the picture management platform and scene types corresponding to the picture data; the training unit is configured to respond to the absence of a picture prediction model corresponding to the scene type, and when judging that the picture data meets the data requirement of model training, the training unit takes the picture data as input and takes a labeling result graph labeled as a target area in the picture data as output to train to obtain the picture prediction model corresponding to the scene type; and a storage unit configured to store the picture prediction model and a correspondence between the picture prediction model and the scene type.
In some embodiments, the apparatus further comprises: and a generation unit configured to input the picture data to the picture prediction model in response to the presence of the picture prediction model corresponding to the scene type, and generate an annotation result map corresponding to the input picture data and the number of target areas in the annotation result map.
In some embodiments, the apparatus further comprises: and the sending unit is configured to send the picture data, the labeling result graph corresponding to the picture data and the number of target areas corresponding to the labeling result to the picture management platform.
In some embodiments, the training unit is further configured to train to obtain a picture prediction model corresponding to the scene type, with the picture data as input and the blob map in the picture data marked as the target connected region as output.
In some embodiments, the network structure of the picture prediction model in the device is built based on the depth residual network and the feature map pyramid network; the depth residual network is resnet-50 and the feature map pyramid network is a recursive FPN.
In some embodiments, the apparatus further comprises: the second sending unit is configured to send prompt information corresponding to the model training result to the picture management platform.
In a fourth aspect, embodiments of the present disclosure provide a terminal device, including: one or more processors; a storage device having one or more programs stored thereon, which when executed by one or more processors, cause the one or more processors to implement the method as described in any of the implementations of the first aspect.
In a fifth aspect, embodiments of the present disclosure provide a computer readable medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements a method as described in any of the implementations of the first aspect.
According to the information processing method, system and device, picture data and scene types corresponding to the picture data are received, the picture prediction model corresponding to the scene types is responded, when the picture data are judged to meet the data requirement of model training, the picture data are taken as input, a labeling result graph labeled as a target area in the picture data is taken as output, the picture prediction model corresponding to the scene types is obtained through training, and the picture prediction model and the corresponding relation between the picture prediction model and the scene types are stored. The method, the system and the equipment for processing the information aiming at the multiple scenes based on the machine learning are realized, and the method, the system and the device for intelligently checking the multiple scenes based on the machine learning are further realized.
It should be understood that the description in this section is not intended to identify key or critical features of the embodiments of the disclosure, nor is it intended to be used to limit the scope of the disclosure. Other features of the present disclosure will become apparent from the following specification.
Drawings
Other features, objects and advantages of the present disclosure will become more apparent upon reading of the detailed description of non-limiting embodiments, made with reference to the following drawings:
FIG. 1 is an exemplary system architecture diagram in which some embodiments of the present disclosure may be applied;
FIG. 2 is a flow chart of one embodiment of an information processing method according to the present disclosure;
FIG. 3 is a schematic diagram of one application scenario of the information processing method according to the present disclosure;
FIG. 4 is a flow chart of yet another embodiment of an information processing method according to the present disclosure;
FIG. 5 is a schematic diagram of an architecture of one embodiment of an information handling system according to the present disclosure;
FIG. 6 is a schematic structural view of one embodiment of an information processing apparatus according to the present disclosure;
Fig. 7 is a schematic structural diagram of an electronic device suitable for use in implementing embodiments of the present disclosure.
Detailed Description
The present disclosure is described in further detail below with reference to the drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and are not limiting of the invention. It should be noted that, for convenience of description, only the portions related to the present invention are shown in the drawings.
It should be noted that, without conflict, the embodiments of the present disclosure and features of the embodiments may be combined with each other. The present disclosure will be described in detail below with reference to the accompanying drawings in conjunction with embodiments.
Fig. 1 illustrates an exemplary system architecture 100 to which an information processing method or an information processing apparatus of an embodiment of the present disclosure may be applied.
As shown in fig. 1, the system architecture 100 may include a picture taking device 101, a picture management platform 102, a model management platform 103, and a terminal device 104.
The model management platform 103 is communicatively connected to the picture management platform 102 by using an interface protocol, where the model management platform 103 may be a server or an electronic device that provides picture prediction and model training, for example, the model management platform 102 determines the picture data sent by the picture management platform 102 and the scene type corresponding to the picture data, performs model training and/or picture prediction according to the determination result, and returns the result of the model training and/or the result of the picture prediction to the picture management platform 102.
The picture management platform 102 is respectively in communication connection with the picture acquisition device 101, the model management platform 103 and the terminal device 104, where the picture management platform may be a server or an electronic device for managing pictures, picture prediction results, and scene types of the picture acquisition device, for example, receives a video stream sent by the picture acquisition device 101, obtains picture data in the video stream, determines a scene type corresponding to the picture data according to scene configuration information of the picture acquisition device 101, sends the picture data and the scene type to the model management platform 103, receives the picture prediction results and the picture data returned by the model management platform 103, integrates the picture prediction results and the picture data, and sends the integrated picture prediction results and the integrated picture data to the terminal device 104. The picture management platform 102 stores configuration files or management files corresponding to pictures, picture prediction results and scene types of the picture acquisition device.
The picture taking device 101 is communicatively connected to the picture management platform 102 using various interface protocols, where the interface protocols may include at least: the software development kit SDK, the open network video interface ONVIF, GB/T28181, and the real time streaming protocol RTSP, the picture taking device 101 may include: the picture capture device 101 may be provided in various mainstream cameras and personalized AI cameras, such as mainstream cameras of various brands, AI edge cameras, and other edge cameras.
The terminal device 104 may be a variety of electronic devices having a display screen and supporting web browsing, including but not limited to smartphones, tablets, electronic book readers, laptop and desktop computers, and the like.
Communication connections between the various devices in the system architecture 100 may be implemented through a network, where the network may be a medium that provides a communication link between the picture taking device 101, the picture management platform 102, the model management platform 103, and the terminal device 104, and may include various connection types, such as a wired, wireless communication link, or an optical fiber cable, etc.
It should be noted that, the information processing method provided in the embodiment of the present application is generally executed by the model management platform 103, and accordingly, the information processing apparatus is generally disposed in the model management platform 103.
It should be understood that the numbers of picture acquisition devices, picture management platforms, model management platforms and terminal devices in fig. 1 are merely illustrative. According to the implementation requirement, the system can be provided with any number of picture acquisition devices, picture management platforms, model management platforms and terminal equipment.
With continued reference to FIG. 2, a flow 200 of one embodiment of an information processing method according to the present disclosure is shown. The information processing method comprises the following steps:
step 201, receiving picture data and a scene type corresponding to the picture data sent by a picture management platform.
In this embodiment, the execution subject (for example, the model management platform shown in fig. 1) may receive, in real time or in a certain period of time, the picture data sent by the picture management platform and the scene type corresponding to the picture data. The picture refers to a planar medium composed of graphics, images, and the like. The scene types can be divided according to the types of the articles, such as metal reinforcing steel bars, metal steel pipes, agricultural and sideline products-pigs, agricultural and sideline products-cattle, energy products-crude oil, rubber and the like, can be divided according to the positions of the articles, such as production rooms, test rooms, warehouse shelves and the like, and can be divided according to the different time points. Scene types may be partitioned according to various application scenes or different states of the application scenes. Here, the number of the picture data may be one or a plurality, and the format of the picture data may be various picture formats, such as bmp format, png format, gif format, JPEG format, AI format, PSD format, etc., and the number of the picture data and the format of the picture data are not further limited.
Step 202, in response to the absence of a picture prediction model corresponding to a scene type, when it is determined that the picture data meets the data requirement of model training, taking the picture data as input and taking a labeling result graph labeled as a target area in the picture data as output, and training to obtain the picture prediction model corresponding to the scene type.
In this embodiment, after receiving the picture data and the scene type in step 201, the executing body may determine whether a picture prediction model corresponding to the scene type exists in the model library, further determine whether the picture data meets the data requirement of model training when determining that the picture data does not exist, and use the picture data as input and the labeling result diagram labeled as the target area in the picture data as output when determining that the picture data meets the data requirement of model training, and train to obtain the picture prediction model corresponding to the scene type by using a machine learning technology.
In some optional implementations of this embodiment, taking the picture data as input and the labeling result graph labeled as the target area in the picture data as output, training to obtain the picture prediction model corresponding to the scene type includes: and taking the picture data as input and the blob map marked as the target communication area in the picture data as output, and training to obtain a picture prediction model corresponding to the scene type. By performing blob analysis on the image, an abnormal communication area is used as a target communication area, and a model prediction method is realized.
In some optional implementations of the present embodiments, the network structure of the picture prediction model is constructed based on a depth residual network and a feature map pyramid network; the depth residual network is resnet-50 and the feature map pyramid network is a recursive FPN. As an example, the feature fusion process of the feature map pyramid network may be: the method comprises the steps of carrying out 2-time up-sampling on a RESNET LAYER down-sampled 32-time feature map and then fusing the feature map with a RESNET LAYER down-sampled 16-time feature map to obtain a first fused feature map, wherein the first fused feature map is a feature map based on the original image after 16-time down-sampling; fusing the up-sampled first fused feature map 2 times with the down-sampled feature map RESNET LAYER times 8 times to obtain a second fused feature map, wherein the second feature map is based on the down-sampled feature map 8 times of the original map; and 8 times of up-sampling is carried out on the second fusion feature map, and a result map corresponding to the original map is obtained. Through combination of feature extraction and feature fusion, FPN is utilized to perform multi-scale feature fusion, so that the extracted information is more abundant, and meanwhile, enough feature information can be obtained for a small target, and a more refined model training method is realized.
Further explaining the picture prediction model, attention mechanisms can be introduced in the feature extraction stage in the picture prediction model training process, so that the model can learn key information, the design of the Attention mechanism can be utilized to strengthen the learning of the coverage range by the inventory algorithm, and the effect of automatic identification can be achieved. The prediction process can be expanded from single-point supervision to multi-point supervision, so that supervised learning of an algorithm is realized. For example, by expanding the labeling information of single-point supervision into one 3*3 core, the effect of increasing the optimized gradient of multi-point supervision is achieved, the size of the core can be designed according to different scenes, when the coverage area of a camera is larger, a 3*3 core can be used when a target object is relatively smaller, the number of cores is expanded to 9, when the coverage area of the camera is smaller, a 5*5 core can be used when the target object is relatively larger, the number of cores is expanded to 25 or more, more effective supervision is achieved, and multi-point supervision can be achieved, so that when pigs are piled up only, and parts of the bodies of the pigs are blocked, inventory can still be accurately carried out.
In some optional implementations of the present embodiment, the method further includes: and stopping executing when the picture data is judged to be not in accordance with the data requirement of the model training. Realizing rapid information processing.
Step 203, storing the picture prediction model and the correspondence between the picture prediction model and the scene type.
In this embodiment, after the training of the picture prediction model in the execution subject confirmation step 202 is completed, the picture prediction model and the correspondence between the picture prediction model and the scene type may be stored in the template library.
In some optional implementations of the present embodiment, the method further includes: and sending prompt information corresponding to the model training result to the picture management platform. After model training is completed, synchronizing the picture prediction model of the scene to a picture management platform so that the next time of calling, the front operation of the model can be directly carried out, and the inventory number of corresponding pictures is calculated.
It should be noted that, a technician may set the model structure of the above-mentioned picture prediction model according to the actual requirement, which is not limited in the embodiments of the present disclosure.
With continued reference to fig. 3, fig. 3 is a schematic diagram 300 of an application scenario of the information processing method according to the present embodiment. The information processing method of the present embodiment operates in the electronic device 301. Firstly, the electronic device 301 receives picture data and a scene type 302 corresponding to the picture data sent by a picture management platform, then the electronic device 301 responds to the absence of a picture prediction model corresponding to the scene type, when judging that the picture data meets the data requirement of model training, the electronic device 301 takes the picture data as input, takes a labeling result graph marked as a target area in the picture data as output, trains to obtain a picture prediction model 303 corresponding to the scene type, and finally the electronic device 301 stores the picture prediction model and a corresponding relation 304 between the picture prediction model and the scene type.
According to the information processing method provided by the embodiment of the disclosure, by receiving the picture data and the scene type corresponding to the picture data sent by the picture management platform, responding to the absence of the picture prediction model corresponding to the scene type, when judging that the picture data meets the data requirement of model training, taking the picture data as input, taking a labeling result graph labeled as a target area in the picture data as output, training to obtain the picture prediction model corresponding to the scene type, and storing the picture prediction model and the corresponding relation between the picture prediction model and the scene type. The method for processing the information aiming at the multiple scenes based on the machine learning is realized, and the intelligent checking method of the multiple scenes based on the machine learning is further realized. The problems of long time consumption, large error and high labor cost of manual checking in the prior art are solved.
With further reference to fig. 4, a flow of yet another embodiment of an information processing method is shown. The flow 400 of the information processing method includes the steps of:
step 401, receiving picture data and a scene type corresponding to the picture data sent by a picture management platform.
In step 402, in response to the existence of the picture prediction model corresponding to the scene type, the picture data is input to the picture prediction model, and the labeling result diagram corresponding to the input picture data and the number of target areas in the labeling result diagram are generated.
In this embodiment, when the execution subject determines that there is a picture prediction model corresponding to a scene type, picture data is input to the picture prediction model, and a labeling result map corresponding to the input picture data and the number of target areas in the labeling result map are generated. The picture prediction model may be pre-trained from historical data. The picture prediction model adopts an end-to-end network model structure, and the input and output sizes are consistent.
Step 403, sending the picture data, the labeling result graph corresponding to the picture data, and the number of the target areas corresponding to the labeling result to the picture management platform.
In this embodiment, the execution subject sends the picture data, the generated labeling result map corresponding to the picture data, and the number of target areas corresponding to the labeling result to the picture management platform.
In this embodiment, the specific operation of step 401 is substantially the same as that of step 201 in the embodiment shown in fig. 2, and will not be described here again.
As can be seen from fig. 4, compared with the embodiment corresponding to fig. 2, the information processing flow 400 in this embodiment inputs the picture data into the picture prediction model in response to the picture prediction model corresponding to the existence of the scene type, generates the labeling result graph corresponding to the input picture data and the number of the target areas in the labeling result graph, and sends the picture data, the labeling result graph corresponding to the picture data, and the number of the target areas corresponding to the labeling result to the picture management platform, so that the data inventory method for different scenes is realized, and the data inventory method has effectiveness in multiple scenes, and can obtain good prediction effects whether for flexible and vivid live pig inventory or for high-density inventory of steel bars, steel pipes and the like.
The terminology used in the present disclosure is for the purpose of describing particular embodiments only and is not intended to be limiting of the disclosure. As used in this disclosure and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It should also be understood that the term "and/or" as used herein refers to and encompasses any or all possible combinations of one or more of the associated listed items. It should be understood that although the terms first, second, third, etc. may be used in this disclosure to describe various information, these information should not be limited to these terms. These terms are only used to distinguish one type of information from another. For example, first information may also be referred to as second information, and similarly, second information may also be referred to as first information, without departing from the scope of the present disclosure. The word "responsive to" as used herein may be interpreted as "at … …" or "when … …", depending on the context.
With further reference to FIG. 5, the present disclosure provides an information handling system, as shown in FIG. 5, comprising: a model management platform 501 and a picture management platform 502, wherein the model management platform is used to execute the information processing method shown above. The picture management platform is used for receiving the video stream sent by the picture acquisition device and acquiring picture data in the video stream; determining a scene type corresponding to the picture data according to the configuration information of the picture acquisition device; and sending the picture data and the scene type to a model management platform.
In the system, the picture management platform is also used for receiving picture data sent by the model management platform, a labeling result graph corresponding to the picture data and the number of target areas corresponding to the labeling result; and sending the received picture data, the labeling result diagram and the number of the target areas to the terminal, so that the terminal displays the picture data, the labeling result diagram and the number of the target areas on a user, and the user can view the result diagram and the original diagram.
In the system, the picture management platform is also used for configuring scene types of the accessed picture acquisition devices; and managing the scene types of the image acquisition devices. The system can configure the acquisition device for different scenes to acquire pictures for the different scenes. By managing the scene types, the prediction of the system is simpler and quicker.
As can be seen from fig. 5, the information processing system realizes a multi-scene picture prediction system based on artificial intelligence and internet of things, and can help each party of bulk commodity futures trade to accurately and timely know the goods storage condition, timely process emergency situations and ensure smooth proceeding of futures trade through automatic goods change identification.
With further reference to fig. 6, as an implementation of the method shown in fig. 2 to 4 described above, the present disclosure provides an embodiment of an information processing apparatus, which corresponds to the method embodiment shown in fig. 2, and which is particularly applicable to various electronic devices.
As shown in fig. 6, the information processing apparatus 600 of the present embodiment includes: a receiving unit 601, a predicting unit 602, and a storing unit 603. The receiving unit is configured to receive the picture data sent by the picture management platform and scene types corresponding to the picture data; the training unit is configured to respond to the absence of a picture prediction model corresponding to the scene type, and when judging that the picture data meets the data requirement of model training, the training unit takes the picture data as input and takes a labeling result graph labeled as a target area in the picture data as output to train to obtain the picture prediction model corresponding to the scene type; and a storage unit configured to store the picture prediction model and a correspondence between the picture prediction model and the scene type.
In this embodiment, the specific processes of the receiving unit 601, the predicting unit 602, and the storing unit 603 of the information processing apparatus 600 and the technical effects thereof may refer to the relevant descriptions of steps 201 to 203 in the corresponding embodiment of fig. 2, and are not repeated here.
In some optional implementations of this embodiment, the apparatus further includes: and a generation unit configured to input the picture data to the picture prediction model in response to the presence of the picture prediction model corresponding to the scene type, and generate an annotation result map corresponding to the input picture data and the number of target areas in the annotation result map.
In some optional implementations of this embodiment, the apparatus further includes: and the sending unit is configured to send the picture data, the labeling result graph corresponding to the picture data and the number of target areas corresponding to the labeling result to the picture management platform.
In some optional implementations of this embodiment, the training unit is further configured to train to obtain a picture prediction model corresponding to the scene type, with the picture data as input and the blob map in the picture data marked as the target connected region as output.
In some optional implementations of this embodiment, the network structure of the picture prediction model in the apparatus is constructed based on the depth residual network and the feature map pyramid network; the depth residual network is resnet-50 and the feature map pyramid network is a recursive FPN.
In some optional implementations of this embodiment, the apparatus further includes: the second sending unit is configured to send prompt information corresponding to the model training result to the picture management platform.
Referring now to fig. 7, a schematic diagram of an electronic device 700 suitable for use in implementing embodiments of the present disclosure is shown. The terminal devices in the embodiments of the present disclosure may include, but are not limited to, mobile terminals such as mobile phones, notebook computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), car terminals (e.g., car navigation terminals), and the like, and stationary terminals such as digital TVs, desktop computers, and the like. The electronic device shown in fig. 7 is only one example and should not impose any limitations on the functionality and scope of use of embodiments of the present disclosure.
As shown in fig. 7, the electronic device 700 may include a processing means (e.g., a central processor, a graphics processor, etc.) 701, which may perform various appropriate actions and processes according to a program stored in a Read Only Memory (ROM) 702 or a program loaded from a storage means 708 into a Random Access Memory (RAM) 703. In the RAM703, various programs and data required for the operation of the electronic device 700 are also stored. The processing device 701, the ROM702, and the RAM703 are connected to each other through a bus 704. An input/output (I/O) interface 705 is also connected to bus 704.
In general, the following devices may be connected to the I/O interface 705: input devices 706 including, for example, a touch screen, touchpad, keyboard, mouse, camera, microphone, accelerometer, gyroscope, and the like; an output device 707 including, for example, a Liquid Crystal Display (LCD), a speaker, a vibrator, and the like; storage 708 including, for example, magnetic tape, hard disk, etc.; and a communication device 709. The communication means 709 may allow the electronic device 700 to communicate wirelessly or by wire with other devices to exchange data. While fig. 7 shows an electronic device 700 having various means, it is to be understood that not all of the illustrated means are required to be implemented or provided. More or fewer devices may be implemented or provided instead. Each block shown in fig. 7 may represent one device or a plurality of devices as needed.
In particular, according to embodiments of the present disclosure, the processes described above with reference to flowcharts may be implemented as computer software programs. For example, embodiments of the present disclosure include a computer program product comprising a computer program embodied on a computer readable medium, the computer program comprising program code for performing the method shown in the flowcharts. In such an embodiment, the computer program may be downloaded and installed from a network via communication device 709, or installed from storage 708, or installed from ROM 702. The above-described functions defined in the methods of the embodiments of the present disclosure are performed when the computer program is executed by the processing device 701.
It should be noted that the computer readable medium of the embodiments of the present disclosure may be a computer readable signal medium or a computer readable storage medium, or any combination of the two. The computer readable storage medium can be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples of the computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In an embodiment of the present disclosure, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. Whereas in embodiments of the present disclosure, the computer-readable signal medium may comprise a data signal propagated in baseband or as part of a carrier wave, with computer-readable program code embodied therein. Such a propagated data signal may take any of a variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination of the foregoing. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: electrical wires, fiber optic cables, RF (radio frequency), and the like, or any suitable combination of the foregoing.
The computer readable medium may be contained in the electronic device; or may exist alone without being incorporated into the electronic device. The computer readable medium carries one or more programs which, when executed by the electronic device, cause the electronic device to: receiving picture data and scene types corresponding to the picture data sent by a picture management platform; responding to the absence of a picture prediction model corresponding to the scene type, when judging that the picture data meets the data requirement of model training, taking the picture data as input, taking a labeling result graph labeled as a target area in the picture data as output, and training to obtain the picture prediction model corresponding to the scene type; and storing the picture prediction model and the corresponding relation between the picture prediction model and the scene type.
Computer program code for carrying out operations of embodiments of the present disclosure may be written in one or more programming languages, including an object oriented programming language such as Java, smalltalk, C ++ and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any kind of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or may be connected to an external computer (for example, through the Internet using an Internet service provider).
The flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
The units involved in the embodiments described in the present disclosure may be implemented by means of software, or may be implemented by means of hardware. The described units may also be provided in a processor, for example, described as: a processor includes a receiving unit, a training unit, and a storage unit. The names of these units do not constitute a limitation on the unit itself in some cases, and for example, the receiving unit may also be described as "a unit that receives the picture data transmitted by the picture management platform and the scene type to which the picture data corresponds".
The foregoing description is only of the preferred embodiments of the present disclosure and description of the principles of the technology being employed. It will be appreciated by those skilled in the art that the scope of the invention in the embodiments of the present disclosure is not limited to the specific combination of the above technical features, but encompasses other technical features formed by any combination of the above technical features or their equivalents without departing from the spirit of the invention. Such as the above-described features, are mutually substituted with (but not limited to) the features having similar functions disclosed in the embodiments of the present disclosure.
Claims (16)
1. An information processing method, comprising:
receiving picture data sent by a picture management platform and scene types corresponding to the picture data;
responding to the absence of a picture prediction model corresponding to the scene type, when judging that the picture data meets the data requirement of model training, taking the picture data as input and taking a labeling result graph labeled as a target area in the picture data as output, and training to obtain the picture prediction model corresponding to the scene type;
Storing the picture prediction model and the corresponding relation between the picture prediction model and the scene type;
The training to obtain the picture prediction model corresponding to the scene type by taking the picture data as input and taking a labeling result diagram labeled as a target area in the picture data as output includes: taking the picture data as input and a blob image marked as a target communication area in the picture data as output, and training to obtain a picture prediction model corresponding to the scene type;
The picture prediction model adopts an end-to-end network model structure, and the input and output sizes are consistent; the picture prediction model is trained based on labeling information of the multi-point supervision, and the size of a kernel corresponding to the multi-point supervision is determined according to the scene type.
2. The method of claim 1, further comprising:
And in response to the existence of the picture prediction model corresponding to the scene type, inputting the picture data into the picture prediction model, and generating an annotation result graph corresponding to the input picture data and the number of target areas in the annotation result graph.
3. The method of claim 2, further comprising:
And sending the picture data, the labeling result graph corresponding to the picture data and the number of target areas corresponding to the labeling result to the picture management platform.
4. The method of claim 1, wherein the network structure of the picture prediction model is constructed based on a depth residual network and a feature map pyramid network; the depth residual network is resnet-50, and the feature map pyramid network is a recursive FPN.
5. The method of claim 1, further comprising:
and sending prompt information corresponding to the model training result to the picture management platform.
6. An information processing system, comprising: model management platform, wherein the model management platform is used for executing the information processing method according to any one of claims 1-5.
7. The system of claim 6, the system further comprising: a picture management platform;
The picture management platform is used for receiving the video stream sent by the picture acquisition device and acquiring picture data in the video stream; determining a scene type corresponding to the picture data according to the configuration information of the picture acquisition device; and sending the picture data and the scene type to the model management platform.
8. The system of claim 7, wherein the picture management platform is further configured to receive picture data sent by the model management platform, a labeling result graph corresponding to the picture data, and a number of target areas corresponding to the labeling result; and sending the received picture data, the labeling result diagram and the number of the target areas to a terminal.
9. The system of claim 7, wherein the picture management platform is further configured to configure a scene type for each of the accessed picture acquisition devices; and managing the scene type of each picture acquisition device.
10. An information processing apparatus comprising:
The receiving unit is configured to receive the picture data sent by the picture management platform and the scene type corresponding to the picture data;
The training unit is configured to respond to the fact that no picture prediction model corresponding to the scene type exists, when the fact that the picture data meets the data requirement of model training is judged, the picture data is taken as input, a labeling result graph marked as a target area in the picture data is taken as output, and the picture prediction model corresponding to the scene type is obtained through training;
A storage unit configured to store the picture prediction model and a correspondence between the picture prediction model and the scene type;
The training unit is further configured to train to obtain a picture prediction model corresponding to the scene type by taking the picture data as input and a blob map marked as a target communication area in the picture data as output;
The picture prediction model adopts an end-to-end network model structure, and the input and output sizes are consistent; the picture prediction model is trained based on labeling information of the multi-point supervision, and the size of a kernel corresponding to the multi-point supervision is determined according to the scene type.
11. The apparatus of claim 10, further comprising:
And the generation unit is configured to input the picture data into the picture prediction model in response to the existence of the picture prediction model corresponding to the scene type, and generate a labeling result diagram corresponding to the input picture data and the number of target areas in the labeling result diagram.
12. The apparatus of claim 11, further comprising:
And the first sending unit is configured to send the picture data, the labeling result graph corresponding to the picture data and the number of target areas corresponding to the labeling result to the picture management platform.
13. The apparatus of claim 10, wherein the network structure of the picture prediction model is constructed based on a depth residual network and a feature map pyramid network; the depth residual network is resnet-50, and the feature map pyramid network is a recursive FPN.
14. The apparatus of claim 10, further comprising:
The second sending unit is configured to send prompt information corresponding to the model training result to the picture management platform.
15. An electronic device, comprising:
One or more processors;
a storage device having one or more programs stored thereon;
when executed by the one or more processors, causes the one or more processors to implement the method of any of claims 1-5.
16. A computer readable medium having stored thereon a computer program, wherein the program when executed by a processor implements the method of any of claims 1-5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111020380.1A CN113643136B (en) | 2021-09-01 | 2021-09-01 | Information processing method, system and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111020380.1A CN113643136B (en) | 2021-09-01 | 2021-09-01 | Information processing method, system and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113643136A CN113643136A (en) | 2021-11-12 |
| CN113643136B true CN113643136B (en) | 2024-06-18 |
Family
ID=78424783
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111020380.1A Active CN113643136B (en) | 2021-09-01 | 2021-09-01 | Information processing method, system and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113643136B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108734162A (en) * | 2018-04-12 | 2018-11-02 | 上海扩博智能技术有限公司 | Target identification method, system, equipment and storage medium in commodity image |
| CN110619423A (en) * | 2019-08-06 | 2019-12-27 | 平安科技（深圳）有限公司 | Multitask prediction method and device, electronic equipment and storage medium |
Family Cites Families (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018176195A1 (en) * | 2017-03-27 | 2018-10-04 | 中国科学院深圳先进技术研究院 | Method and device for classifying indoor scene |
| CN108549893B (en) * | 2018-04-04 | 2020-03-31 | 华中科技大学 | End-to-end identification method for scene text with any shape |
| CN110443969B (en) * | 2018-05-03 | 2021-06-04 | 中移(苏州)软件技术有限公司 | Fire detection method and device, electronic equipment and storage medium |
| CN109002842A (en) * | 2018-06-27 | 2018-12-14 | 北京字节跳动网络技术有限公司 | Image-recognizing method and device |
| US10304193B1 (en) * | 2018-08-17 | 2019-05-28 | 12 Sigma Technologies | Image segmentation and object detection using fully convolutional neural network |
| CN109919244B (en) * | 2019-03-18 | 2021-09-07 | 北京字节跳动网络技术有限公司 | Method and apparatus for generating a scene recognition model |
| CN111723746A (en) * | 2020-06-22 | 2020-09-29 | 江苏云从曦和人工智能有限公司 | Scene recognition model generation method, system, platform, device and medium |
| CN111767371B (en) * | 2020-06-28 | 2024-03-12 | 微医云(杭州)控股有限公司 | Intelligent question-answering method, device, equipment and medium |
| CN112001884A (en) * | 2020-07-14 | 2020-11-27 | 浙江大华技术股份有限公司 | Training method, counting method, equipment and storage medium of quantity statistical model |
| CN111968081B (en) * | 2020-07-30 | 2023-11-21 | 中国农业大学 | Automatic fish shoal counting method and device, electronic equipment and storage medium |
| CN112183303A (en) * | 2020-09-24 | 2021-01-05 | 南方电网数字电网研究院有限公司 | Transformer equipment image classification method and device, computer equipment and medium |
| CN112862849B (en) * | 2021-01-27 | 2022-12-27 | 四川农业大学 | Image segmentation and full convolution neural network-based field rice ear counting method |
| CN113192065A (en) * | 2021-05-28 | 2021-07-30 | 广东电网有限责任公司 | Processing method of power transmission line image, portable equipment and user terminal |
-
2021
- 2021-09-01 CN CN202111020380.1A patent/CN113643136B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108734162A (en) * | 2018-04-12 | 2018-11-02 | 上海扩博智能技术有限公司 | Target identification method, system, equipment and storage medium in commodity image |
| CN110619423A (en) * | 2019-08-06 | 2019-12-27 | 平安科技（深圳）有限公司 | Multitask prediction method and device, electronic equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113643136A (en) | 2021-11-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108710885B (en) | Target object detection method and device | |
| CN109740018B (en) | Method and device for generating video label model | |
| EP3637310A1 (en) | Method and apparatus for generating vehicle damage information | |
| CN109947989B (en) | Method and apparatus for processing video | |
| CN112668588B (en) | Parking space information generation method, device, equipment and computer readable medium | |
| CN110399847B (en) | Key frame extraction method and device and electronic equipment | |
| CN111737573A (en) | Resource recommendation method, device, equipment and storage medium | |
| CN110059623B (en) | Method and apparatus for generating information | |
| CN109816023B (en) | Method and device for generating picture label model | |
| CN113112497B (en) | Industrial appearance defect detection method based on zero sample learning, electronic equipment and storage medium | |
| CN111597465A (en) | Display method and device and electronic equipment | |
| CN112712036A (en) | Traffic sign recognition method and device, electronic equipment and computer storage medium | |
| CN110555861B (en) | Optical flow calculation method and device and electronic equipment | |
| CN110287817B (en) | Target recognition and target recognition model training method and device and electronic equipment | |
| CN111310595B (en) | Method and device for generating information | |
| CN113643136B (en) | Information processing method, system and device | |
| CN110414625B (en) | Method and device for determining similar data, electronic equipment and storage medium | |
| CN113255819B (en) | Method and device for identifying information | |
| CN111488928B (en) | Method and device for acquiring samples | |
| CN113222050B (en) | Image classification method and device, readable medium and electronic equipment | |
| US11610328B2 (en) | Method and apparatus for identifying item | |
| CN116704593A (en) | Predictive model training method, apparatus, electronic device, and computer-readable medium | |
| CN110263743B (en) | Method and device for recognizing images | |
| CN113920397A (en) | Method and device for training image classification model and method and device for image classification | |
| CN110991312A (en) | Method, apparatus, electronic device, and medium for generating detection information |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |