Image processing method, device and electronic equipment

Abstract

The present application provides an image processing method, comprising: obtaining an image to be processed, wherein the image to be processed includes at least two target objects; based on the image to be processed, identifying multiple part areas corresponding to the target objects to obtain multiple sub-images, wherein the multiple sub-images contain characteristic information of the target objects; associating multiple sub-images corresponding to the same target object with the target object to obtain multiple sub-images corresponding to the target object; based on the characteristic information of the target object, performing recognition processing on the multiple sub-images corresponding to the target object to obtain a recognition result for the target object. When analyzing an image to obtain a recognition result of a target object in the image, the present application associates multiple sub-images corresponding to the same target object in the image with the target object, so that the subsequent method of identifying the target object is relatively simple. This solves the problem that the existing method of identifying the target object in the image is relatively complicated.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to an image processing method, device, electronic device and computer storage medium. Background Art

[0002] With the continuous development of digital technology, the application of video equipment in daily life is becoming more and more extensive. As a result, more and more video recording content is generated. As we all know, video recording content is mainly used for people to query useful information about the monitoring environment. For example, in home intelligent monitoring, you can find the video segment information of the target object or object of interest in the camera video from the video recording content. Or in public environment monitoring, find the video segment information of the target object or object of interest in the real-time video recording content, so as to track the movement trajectory information of the target object or object of interest.

[0003] After obtaining the video recording content, the target object is generally identified based on the video frame. If multiple parts of at least one target object in the video frame need to be identified, most of the existing technologies identify each part of the target object separately. However, since this identification method identifies multiple parts separately, when there are multiple target objects in the video frame, it is necessary to correspond the identification results of different parts with the target object, resulting in the identification process being not simple enough. Summary of the invention

[0004] The present application provides an image processing method to solve the problem that the existing method of identifying a target object in a video frame is relatively complicated. The present application also provides an image processing device, and corresponding electronic equipment and computer media.

[0005] The present application provides an image processing method, including:

[0006] Obtaining an image to be processed, wherein the image to be processed includes at least two target objects;

[0007] Based on the image to be processed, identifying multiple part regions corresponding to the target object, and obtaining multiple sub-images, wherein the multiple sub-images contain feature information of the target object;

[0008] Associating multiple sub-images corresponding to the same target object with the target object to obtain multiple sub-images corresponding to the target object;

[0009] Based on the feature information of the target object, recognition processing is performed on the multiple sub-images corresponding to the target object to obtain a recognition result for the target object.

[0010] Optionally, performing recognition processing on the multiple sub-images corresponding to the target object based on the feature information of the target object to obtain a recognition result for the target object includes:

[0011] Based on the feature information of the target object, respectively perform recognition processing on the multiple sub-images corresponding to the target object to obtain multiple recognition results;

[0012] According to the multiple recognition results, a recognition result for the target object is obtained.

[0013] Optionally, the image to be processed refers to a video frame image;

[0014] The video frame image is obtained by:

[0015] Sending a request to the video acquisition device to obtain the video file;

[0016] The video file sent by the video acquisition device is received, and a video frame image is obtained according to the video file.

[0017] Optionally, associating multiple sub-images corresponding to the same target object with the target object to obtain multiple sub-images corresponding to the target object includes:

[0018] Get the current target object;

[0019] Based on the feature information of the target object contained in the sub-image, determining a plurality of sub-images matching the current target object;

[0020] Establish an association relationship between the multiple sub-images matching the current target object and the current target object to obtain multiple sub-images corresponding to the target object.

[0021] Optionally, the determining a plurality of sub-images matching the current target object based on feature information of the target object contained in the sub-images includes:

[0022] Based on the feature information of the target object contained in the sub-image, traverse each sub-image of the multiple sub-images to obtain the matching degree information between each sub-image and the current target object;

[0023] According to the matching degree information between each sub-image and the current target object and a preset matching degree threshold condition, a plurality of sub-images matching the current target object are determined.

[0024] Optionally, determining a plurality of sub-images matching the current target object according to the matching degree information between each sub-image and the current target object and a matching degree threshold condition includes:

[0025] Determine whether the matching degree information between the sub-image and the current target object meets the matching degree threshold condition;

[0026] If the sub-image meets the matching degree threshold condition, the sub-image is determined as a sub-image matching the current target object; in the same manner, multiple sub-images matching the current target object are determined.

[0027] Optionally, the identifying, based on the image to be processed, a plurality of part regions corresponding to the target object to obtain a plurality of sub-images includes:

[0028] The image to be processed is used as input data of a convolutional neural network and an image region candidate frame generation network to obtain a partition result of the image to be processed; the convolutional neural network and the image region candidate frame generation network are used together as a neural network for partitioning the image to be processed to obtain an image partition result;

[0029] A plurality of sub-images are obtained according to the partitioning result of the image to be processed.

[0030] Optionally, the multiple sub-images include a first sub-image and a second sub-image;

[0031] The step of performing recognition processing on the multiple sub-images corresponding to the target object based on the feature information of the target object to obtain multiple recognition results includes:

[0032] Performing quality assessment on the first sub-image and the second sub-image, and performing first key point positioning and second key point positioning on feature information included in the first sub-image and feature information included in the second sub-image, respectively, to obtain first key feature information and second key feature information that meet preset image quality assessment conditions;

[0033] The first key feature information and the second key feature information that meet the preset image quality assessment condition are identified to obtain a first identification result and the second identification result, and the first identification result and the second identification result are confirmed as the multiple identification results.

[0034] Optionally, the multiple sub-images further include a third sub-image;

[0035] The step of performing recognition processing on the multiple sub-images corresponding to the target object based on the feature information of the target object to obtain multiple recognition results includes:

[0036] Performing third key point positioning on the feature information contained in the third sub-image to obtain third key feature information;

[0037] Classifying the feature information contained in the third sub-image to obtain a category of the feature information of the third sub-image;

[0038] Based on the third key feature information and the feature information category of the third sub-image, a third recognition result is obtained, and the first recognition result, the second recognition result, and the third recognition result are confirmed as the multiple recognition results.

[0039] Optionally, it further includes: determining whether the target objects overlap;

[0040] If the target objects overlap, segmentation processing is performed on the image to be processed.

[0041] Optionally, it also includes: determining whether the target object in the current image meets a preset condition;

[0042] If the target object in the current image does not meet the preset condition, an image in which the target object meets the preset condition is obtained again, and the image in which the target object meets the preset condition is used as the image to be processed.

[0043] Optionally, the method further includes: displaying the recognition result for the target object and obtaining feedback information for the recognition result;

[0044] Based on the feedback information, it is determined whether to re-identify the recognition result.

[0045] Correspondingly, an embodiment of the present application provides an image processing device, including:

[0046] A unit for obtaining an image to be processed, used for obtaining an image to be processed, wherein the image to be processed includes at least two target objects;

[0047] A sub-image obtaining unit, configured to identify a plurality of part regions corresponding to the target object based on the image to be processed, and obtain a plurality of sub-images, wherein the plurality of sub-images contain feature information of the target object;

[0048] an associating unit, used to associate multiple sub-images corresponding to the same target object with the target object to obtain multiple sub-images corresponding to the target object;

[0049] The target object recognition unit is used to perform recognition processing on the multiple sub-images corresponding to the target object based on the feature information of the target object to obtain a recognition result for the target object.

[0050] The present application provides an image processing method, including:

[0051] Obtaining an image to be processed, wherein the image to be processed includes a target object;

[0052] Based on the image to be processed, identifying the image to be processed, and obtaining a plurality of sub-images, wherein the plurality of sub-images contain feature information of a target object;

[0053] Associating multiple sub-images corresponding to the same target object to obtain multiple sub-images corresponding to the target object;

[0054] Based on the feature information of the target object, recognition processing is performed on the multiple sub-images corresponding to the target object to obtain a recognition result for the target object.

[0055] The present application provides a traffic image processing method, including:

[0056] Obtaining a traffic image to be processed, wherein the traffic image to be processed includes vehicles;

[0057] Based on the traffic image to be processed, identifying the traffic image to be processed, and obtaining a plurality of sub-images, wherein the plurality of sub-images contain characteristic information of vehicles;

[0058] Associating multiple sub-images corresponding to the same vehicle to obtain multiple sub-images corresponding to the vehicle;

[0059] Based on the characteristic information of the vehicle, recognition processing is performed on the multiple sub-images corresponding to the vehicle to obtain a recognition result for the vehicle.

[0060] The embodiment of the present application provides an image processing device, including: a collection device, a recognition result acquisition device;

[0061] The acquisition device is used to obtain an image to be processed, wherein the image to be processed includes a target object;

[0062] The recognition result acquisition device is used to identify the image to be processed based on the image to be processed to obtain multiple sub-images, wherein the multiple sub-images contain characteristic information of the target object; associate multiple sub-images corresponding to the same target object to obtain multiple sub-images corresponding to the target object; and perform recognition processing on the multiple sub-images corresponding to the target object based on the characteristic information of the target object to obtain a recognition result for the target object.

[0063] Optionally, a display device is also included;

[0064] The display device is used to display the recognition result for the target object.

[0065] Correspondingly, an embodiment of the present application provides an electronic device, including:

[0066] processor;

[0067] The memory is used to store a computer program, the computer program is executed by the processor to perform an image processing method, the method comprising the following steps:

[0068] Obtaining an image to be processed, wherein the image to be processed includes at least two target objects;

[0069] Based on the image to be processed, identifying multiple part regions corresponding to the target object, and obtaining multiple sub-images, wherein the multiple sub-images contain feature information of the target object;

[0070] Associating multiple sub-images corresponding to the same target object with the target object to obtain multiple sub-images corresponding to the target object;

[0071] Based on the feature information of the target object, recognition processing is performed on the multiple sub-images corresponding to the target object to obtain a recognition result for the target object.

[0072] Correspondingly, an embodiment of the present application provides a computer storage medium, wherein the computer storage medium stores a computer program, and the computer program is executed by a processor to perform an image processing method, wherein the method comprises the following steps:

[0073] Obtaining an image to be processed, wherein the image to be processed includes at least two target objects;

[0074] Based on the image to be processed, identifying multiple part regions corresponding to the target object, and obtaining multiple sub-images, wherein the multiple sub-images contain feature information of the target object;

[0075] Associating multiple sub-images corresponding to the same target object with the target object to obtain multiple sub-images corresponding to the target object;

[0076] Based on the feature information of the target object, recognition processing is performed on the multiple sub-images corresponding to the target object to obtain a recognition result for the target object.

[0077] Compared with the prior art, this application has the following advantages:

[0078] The embodiment of the present application provides an image processing method, comprising: obtaining an image to be processed, wherein the image to be processed includes at least two target objects; based on the image to be processed, identifying multiple part areas corresponding to the target objects to obtain multiple sub-images, wherein the multiple sub-images contain characteristic information of the target objects; associating multiple sub-images corresponding to the same target object with the target object to obtain multiple sub-images corresponding to the target object; based on the characteristic information of the target object, performing recognition processing on the multiple sub-images corresponding to the target object to obtain a recognition result for the target object. When analyzing an image including a target object to obtain a recognition result of the target object in the image, the present application associates multiple sub-images corresponding to the same target object in the image with the target object for the target object in the image to obtain multiple sub-images corresponding to the target object, so that when a target object is recognized, multiple characteristic information of the target object is matched with the target object, so that the subsequent method of identifying the target object is relatively simple. The problem that the existing method of identifying the target object in the image is relatively complicated is solved. BRIEF DESCRIPTION OF THE DRAWINGS

[0079] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present application. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0080] Figure 1-A This is a first schematic diagram of an application scenario of the image processing method provided in this application.

[0081] Figure 1-B This is a second schematic diagram of an application scenario of the image processing method provided in this application.

[0082] Figure 2 This is a flow chart of the image processing method provided in the first embodiment of the present application.

[0083] Figure 3 This is a schematic diagram of the behavior recognition algorithm framework of the first embodiment of the present application.

[0084] Figure 4 A schematic diagram of an image processing device provided in the second embodiment of the present application.

[0085] Figure 5 This is a flow chart of the image processing method provided in the third embodiment of the present application.

[0086] Figure 6 This is a flow chart of the traffic image processing method provided in the fourth embodiment of the present application.

[0087] Figure 7 A schematic diagram of an image processing electronic device provided in the sixth embodiment of the present application. DETAILED DESCRIPTION

[0088] Many specific details are described in the following description to facilitate a full understanding of the present application. However, the present application can be implemented in many other ways than those described herein, and those skilled in the art can make similar generalizations without violating the connotation of the present application. Therefore, the present application is not limited to the specific implementation disclosed below.

[0089] In order to more clearly demonstrate the image processing method provided by the present application, the application scenario of the image processing method provided by the present application is first introduced. The image processing method provided by the present application can be applied to animal identification in a zoo to help animal managers or breeders manage or raise animals.

[0090] like Figure 1-A and Figure 1-B As shown, it is a schematic diagram of the application scenario of the image processing method provided by this application. Figure 1-A and Figure 1-B The application scenario of the image processing method is described. First, an image to be processed is obtained. The image to be processed in this application may refer to an image to be recognized. Figure 1-A and Figure 1-B The image to be processed is represented by the image to be analyzed. The image to be identified can be a video frame based on a video file shot by a video camera. A video camera can refer to a camera installed in a certain space to collect real-time behavior data of the target object. For example, in the monkey house of a certain zoo, in order to observe the daily physical health status and eating habits of multiple monkeys, cameras are installed in the monkey house. The number of cameras only needs to ensure that all the monkeys in the monkey house can be collected. The camera equipment can collect audio data and image data of all the monkeys in the monkey house. In this scenario, the target objects are all the monkeys in the monkey house. After obtaining the video file of the camera device, the video file is framed to obtain video frames, that is, the image to be identified. Figure 1-AAs shown, the image to be identified contains two monkeys. Of course, the image containing multiple monkeys or the image containing one monkey can also be used as the image to be identified. After obtaining the image to be identified, the image to be identified is subjected to regional identification to obtain multiple sub-images. Specifically, the region including the monkey face, head and upper body in the image to be identified can be identified and marked to obtain multiple sub-images. Meanwhile, multiple sub-images can also be classified further, for example, the sub-image corresponding to the monkey face can be used as the first sub-image, the sub-image corresponding to the monkey head can be used as the second sub-image, and the sub-image corresponding to the monkey upper body can be used as the third sub-image. Here, the monkey head can include the monkey face, and the monkey upper body can include the monkey head and the monkey face.

[0091] After obtaining multiple sub-images, the sub-images are associated with the target objects corresponding to the sub-images. That is, each sub-image corresponds to a monkey, and the sub-images are associated with the monkey. Specifically, multiple sub-images corresponding to the same monkey are actually associated with the monkey. Figure 1-A As shown, through the image to be identified, the first sub-image of the first monkey, the second sub-image of the first monkey, and the third sub-image of the first monkey are obtained. At the same time, the first sub-image of the second monkey, the second sub-image of the second monkey, and the third sub-image of the second monkey can also be obtained. The first sub-image of the first monkey, the second sub-image of the first monkey, and the third sub-image of the first monkey are all sub-images associated with the first monkey. Similarly, the first sub-image of the second monkey, the second sub-image of the second monkey, and the third sub-image of the second monkey are all sub-images associated with the second monkey.

[0092] Then, multiple sub-images associated with each monkey are identified. Specifically, taking the identification of a monkey in an image as an example, the identification is based on the characteristic information of the monkey in the sub-image, and multiple sub-images corresponding to the monkey are identified and processed to obtain the identification result for the monkey. More specifically, the identification is performed on multiple sub-images corresponding to the monkey to obtain the identification result for the monkey, which means: first, based on the characteristic information of the monkey, multiple sub-images corresponding to the monkey are identified and processed respectively to obtain multiple identification results. Then, based on the multiple identification results, the identification result for the monkey is obtained.

[0093] Specifically, based on the characteristic information of the monkey, multiple sub-images corresponding to the monkey are respectively identified and processed to obtain multiple identification results, in the following manner:

[0094] First, quality assessment is performed on the first sub-image corresponding to the monkey and the second sub-image corresponding to the monkey, and first key point positioning and second key point positioning are performed on the feature information contained in the first sub-image corresponding to the monkey and the feature information contained in the second sub-image corresponding to the monkey, respectively, to obtain first key feature information and second key feature information that meet preset image quality assessment conditions.

[0095] Afterwards, the first key feature information and the second key feature information that meet the preset image quality assessment condition are identified to obtain a first identification result and a second identification result.

[0096] The third sub-image is identified while the first sub-image is identified. The process of identifying the third sub-image is described as follows: first, the third key point of the feature information contained in the third sub-image corresponding to the monkey is located to obtain the third key feature information, and the feature information contained in the third sub-image corresponding to the monkey is classified to obtain the category of the feature information of the third sub-image.

[0097] Afterwards, a third recognition result is obtained based on the third key feature information and the feature information category of the third sub-image. Finally, the first recognition result, the second recognition result and the third recognition result are confirmed as a plurality of recognition results.

[0098] The first recognition result may be the monkey's expression result based on the monkey's facial feature information, the second recognition result may be the monkey's identity result based on the monkey's head feature information, and the third recognition result may be the monkey's action result based on the monkey's upper body feature information. Figure 1-A As shown, taking the first monkey as an example, through the first sub-image of the first monkey, the second sub-image of the first monkey and the third sub-image of the first monkey, the first recognition result of the first monkey, the second recognition result of the first monkey and the third recognition result of the first monkey can be obtained, and the first recognition result of the first monkey, the second recognition result of the first monkey and the third recognition result of the first monkey are collectively used as the recognition result of the first monkey. Similarly, through the first sub-image of the second monkey, the second sub-image of the second monkey and the third sub-image of the second monkey, the first recognition result of the second monkey, the second recognition result of the second monkey and the third recognition result of the second monkey can be obtained, and the first recognition result of the second monkey, the second recognition result of the second monkey and the third recognition result of the second monkey are collectively used as the recognition result of the second monkey. Of course, no matter how many target objects are contained in the image to be recognized, according to the following Figure 1-A and Figure 1-B The recognition result corresponding to the target object can be obtained in this way.

[0099] The application scenarios of the above-mentioned image processing methods are merely embodiments of the application scenarios of the image processing methods provided by the present application. The purpose of providing the application scenario embodiments is to facilitate understanding of the image processing methods provided by the present application, and is not intended to limit the image processing methods provided by the present application. Other application scenarios of the image processing methods in the embodiments of the present application are not described one by one.

[0100] The present application provides an image processing method, device, electronic device and computer storage medium, and the following are specific embodiments.

[0101] like Figure 2 As shown, it is a flow chart of an image processing method embodiment provided by the first embodiment of the present application. The method comprises the following steps.

[0102] Step S201: obtaining an image to be processed, wherein the image to be processed includes at least two target objects.

[0103] As the first step of the image processing method of the first embodiment, firstly, an image to be processed is obtained, wherein the image to be processed contains a target object. The image to be processed may refer to a video frame image. The image to be processed may be obtained in the following manner: first, a request to obtain a video file is sent to a video acquisition device. Then, a video file sent by the video acquisition device is received, and a video frame image is obtained according to the video file.

[0104] Specifically, taking the application of the image processing method in the scene of the zoo as an example, the image to be identified can be a video frame of a video file shot by a video camera. The video camera can refer to a camera installed in a certain space to collect real-time behavior data of the target object. For example, in a monkey house in a zoo, in order to observe the daily physical health status and eating habits of multiple monkeys, a camera is installed in the monkey house. The number of cameras only needs to ensure that all the monkeys in the monkey house can be collected, and the camera can collect audio data and image data of all the monkeys in the monkey house. In this scenario, the target object is all the monkeys in the monkey house. After obtaining the video file of the camera, the video file is framed to obtain a video frame, that is, the image to be identified.

[0105] In order to send the physical health status and eating habits of all the monkeys in the monkey house to the keeper's client in time, real-time video monitoring of the multiple monkeys in the monkey house is required. Multiple cameras can be installed at multiple locations in the monkey house to ensure that multiple cameras can monitor all the monkeys in the monkey house.

[0106] After the multiple camera devices collect the image data and audio data in the monkey house for a period of time, the multiple video files saved by the multiple camera devices are sent to the server. Specifically, the keeper may use the client to send a request to obtain the video file to each of the multiple camera devices, and each camera device, after receiving the request, sends its own video file to the client or the server, thereby further achieving the acquisition of the image to be processed in step S201.

[0107] When multiple video files are sent to the client, the client can send the multiple video files to the server. When multiple video files are sent to the server, the server can directly obtain the multiple video files. Regardless of which method is adopted, the multiple video files need to be finally sent to the server for processing the video files.

[0108] Step S202: Based on the image to be processed, identify multiple part regions corresponding to the target object to obtain multiple sub-images, wherein the multiple sub-images contain feature information of the target object.

[0109] After the image to be processed is obtained in step S201, multiple part regions corresponding to the target object in the image to be processed are identified to obtain multiple sub-images, wherein the multiple sub-images contain feature information of the target object.

[0110] Specifically, identifying multiple part regions corresponding to the target object in the image to be processed and obtaining multiple sub-images is actually performing partition recognition on the image to be recognized and obtaining multiple sub-images based on the partition recognition results.

[0111] As an implementation method for identifying multiple part regions corresponding to the target object in the image to be processed and obtaining multiple sub-images, the image to be processed may be used as input data of a convolutional neural network and an image region candidate frame generation network to obtain a partition result of the image to be processed. The convolutional neural network and the image region candidate frame generation network are used together as a neural network for partitioning the image to be processed to obtain an image partition result.

[0112] The above-mentioned convolutional neural network can extract the characteristic information of the target object in the image, and the image region candidate frame generation network can partition the image to be identified based on the characteristic information of the target object extracted by the convolutional neural network to obtain a partition result. For example, in an image to be identified containing multiple monkeys, a convolutional neural network can be used to obtain the characteristic information of the monkeys in the image, and obtain the facial characteristic information, head characteristic information and upper body characteristic information of the monkeys. After obtaining the characteristic information of the target object, the image region candidate frame generation network can partition the image to be identified according to the characteristic information of the target object to obtain a partition result. The partition result refers to the partition result of the image to be identified. After obtaining the partition result of the image to be identified, multiple sub-images are obtained according to the partition result of the image to be identified. For example, after identifying the faces, heads and upper bodies of multiple monkeys in the image to be identified, the identified faces, heads and upper bodies are marked in different ways to obtain multiple sub-images.

[0113] Step S203: Associating multiple sub-images corresponding to the same target object with the target object to obtain multiple sub-images corresponding to the target object.

[0114] After multiple part regions corresponding to the target object are identified in step S202 and multiple sub-images are obtained, for each target object, multiple sub-images corresponding to the same target object are associated with the target object to obtain multiple sub-images corresponding to the target object.

[0115] Specifically, as a way to associate multiple sub-images corresponding to the same target object with the target object to obtain multiple sub-images corresponding to the target object: first, obtain the current target object. Then, based on the feature information of the target object contained in the sub-image, determine multiple sub-images matching the current target object. Finally, establish an association relationship between the multiple sub-images matching the current target object and the current target object to obtain multiple sub-images corresponding to the target object.

[0116] For example, when the image to be identified includes two monkeys, if only the face, head and upper body of the monkeys in the image are identified, six sub-images are obtained, namely, the sub-image corresponding to the face of the first monkey, the sub-image corresponding to the face of the second monkey, the sub-image corresponding to the head of the first monkey, the sub-image corresponding to the head of the second monkey, the sub-image corresponding to the upper body of the first monkey, and the sub-image corresponding to the upper body of the second monkey.

[0117] For each monkey, for example, monkey A among the two monkeys, multiple sub-images corresponding to monkey A are associated with monkey A to obtain multiple sub-images corresponding to monkey A. Specifically, the process is to determine multiple sub-images matching monkey A based on the characteristic information of the monkey contained in the sub-image. Determining multiple sub-images matching monkey A based on the characteristic information of the monkey contained in the sub-image may be performed in the following manner.

[0118] First, based on the characteristic information of the monkey contained in the sub-image, each of the six sub-images is traversed to obtain the matching degree information between each sub-image and monkey A. Of course, all characteristic information of monkey A may be pre-stored in a database, and the characteristic information of the monkeys in the multiple sub-images is matched with the characteristic information of monkey A stored in the database.

[0119] Afterwards, multiple sub-images matching monkey A are determined based on the matching degree information between each sub-image and monkey A and a preset matching degree threshold condition.

[0120] As an implementation method of determining multiple sub-images matching monkey A based on the matching degree information between each sub-image and monkey A and the preset matching degree threshold condition, it is determined whether the matching degree information between the sub-image and monkey A meets the matching degree threshold condition. If the sub-image meets the matching degree threshold condition, the sub-image is determined as a sub-image matching monkey A, and multiple sub-images matching monkey A are determined in the same manner.

[0121] For example, among the six sub-images mentioned above, by calculating the matching degree, the sub-image corresponding to the face of the first monkey, the sub-image corresponding to the head of the first monkey, and the sub-image corresponding to the upper body of the first monkey all meet the matching degree threshold condition with monkey A. At this time, the sub-image corresponding to the face of the first monkey, the sub-image corresponding to the head of the first monkey, and the sub-image corresponding to the upper body of the first monkey are all determined as multiple sub-images that match monkey A. The sub-image corresponding to the face of the second monkey, the sub-image corresponding to the head of the second monkey, and the sub-image corresponding to the upper body of the second monkey are all multiple sub-images that do not match monkey A.

[0122] Similarly, when monkey B is the current target object, by calculating the matching degree, the matching degree of the sub-image corresponding to the face of the first monkey, the sub-image corresponding to the head of the first monkey, and the sub-image corresponding to the upper body of the first monkey with monkey B does not meet the matching degree threshold condition. Then the sub-image corresponding to the face of the first monkey, the sub-image corresponding to the head of the first monkey, and the sub-image corresponding to the upper body of the first monkey are all determined as multiple sub-images that do not match monkey B. The sub-image corresponding to the face of the second monkey, the sub-image corresponding to the head of the second monkey, and the sub-image corresponding to the upper body of the second monkey are all multiple sub-images that match monkey B.

[0123] Thus, the above method can achieve that for each target object, multiple sub-images corresponding to the same target object are associated with the target object to obtain multiple sub-images corresponding to the target object.

[0124] Step S204: Based on the feature information of the target object, a plurality of sub-images corresponding to the target object are recognized to obtain a recognition result for the target object.

[0125] After obtaining a plurality of sub-images corresponding to each target object in step S203, recognition processing is performed on the plurality of sub-images corresponding to each target object based on the feature information of the target object to obtain a recognition result for the target object.

[0126] Specifically, based on the feature information of the target object, for each target object, recognition processing is performed on a plurality of sub-images corresponding to the target object, and the recognition result for the target object may be obtained in the manner described below.

[0127] First, based on the feature information of the target object, multiple sub-images corresponding to the target object are respectively recognized to obtain multiple recognition results.

[0128] Since the multiple sub-images corresponding to each target object include the first sub-image and the second sub-image, the first sub-image and the second sub-image corresponding to the target object may be identified in the following manner.

[0129] Based on the feature information of the target object, the first sub-image and the second sub-image corresponding to the target object are respectively recognized to obtain a first recognition result and a second recognition result corresponding to the target object.

[0130] As a method for identifying the first sub-image and the second sub-image corresponding to the target object based on the feature information of the target object, and obtaining the first recognition result and the second recognition result corresponding to the target object, the following implementation method is used: first, the quality of the first sub-image and the second sub-image is evaluated, and the first key point positioning and the second key point positioning are performed on the feature information contained in the first sub-image and the feature information contained in the second sub-image, respectively, to obtain the first key feature information and the second key feature information that meet the preset image quality evaluation conditions. After that, the first key feature information and the second key feature information that meet the preset image quality evaluation conditions are identified to obtain the first recognition result and the second recognition result, and the first recognition result and the second recognition result are confirmed as multiple recognition results.

[0131] For example, when identifying a monkey's face sub-image and head sub-image, the monkey's expression recognition result and identity recognition result can be obtained through the monkey's face sub-image and head sub-image. In practice, after obtaining the monkey's face sub-image and head sub-image, the monkey's face sub-image and head sub-image are firstly quality evaluated. By using the quality evaluation method, images that are not suitable for expression recognition or identity recognition can be filtered out. For example, some relatively blurred images are not suitable for expression recognition or identity recognition. By using the quality evaluation method, these images can be filtered out in advance to improve the accuracy of recognition.

[0132] When performing expression recognition or identity recognition, the key point information in the sub-image is used for recognition, that is, the key feature information is used for recognition. Therefore, after quality assessment, the first key feature information and the second key feature information are obtained, and then recognition is performed based on the first key feature information and the second key feature information. For example, when performing expression recognition, the first key feature information may include key feature information such as eyes and mouth, and when performing identity recognition, the second key feature information may include key feature information such as eyes, nose, ears, and mouth. The expression recognition result of the monkey can be obtained through the first key feature information of the face, such as the expression of the monkey being happy or unhappy or hungry. The identity recognition result of the monkey can be obtained through the second key feature information of the head, such as whether the monkey is monkey A or monkey B.

[0133] Since the multiple sub-images corresponding to each target object include the third sub-image, the third sub-image corresponding to the target object may be identified in the following manner.

[0134] Based on the feature information of the target object, the third sub-image corresponding to the target object is recognized to obtain a third recognition result corresponding to the target object.

[0135] As an implementation method of identifying the third sub-image corresponding to the target object based on the feature information of the target object and obtaining the third recognition result corresponding to the target object: first, the third key point of the feature information contained in the third sub-image is located to obtain the third key feature information. At the same time, the feature information contained in the third sub-image is classified to obtain the category of the feature information of the third sub-image, that is, action classification is performed. Finally, based on the third key feature information and the category of the feature information of the third sub-image, the third recognition result is obtained.

[0136] This embodiment can simultaneously identify the behavior of each target object of multiple target objects in multiple consecutive video frames, such as Figure 3 As shown, it shows a schematic diagram of an algorithm framework for identifying the behavior of each target object in multiple target objects in multiple continuous video frames. The recognition method can be used in the scene of simultaneously identifying the behavior of the upper body of each monkey in multiple monkeys. In practice, the behavior of each target object in multiple target objects in multiple continuous video frames is simultaneously identified based on video files. First, multiple continuous video frames are input into a convolutional neural network and a region candidate frame generation network to obtain feature maps of multiple continuous video frames, and then the target similarity can be obtained by aligning the regional features and matching the target similarity. In this way, the same area to be identified in multiple continuous video frames can be identified, for example, the behavior of the upper body of each monkey in multiple monkeys in multiple continuous video frames can be identified. That is, the algorithm obtains not only the static behavior of a certain monkey, but also the dynamic action behavior of the upper body of a certain monkey in continuous video frames. In the process of identification, features are extracted by a feature extractor, and features of multiple video frames are fused to obtain the action classification of a certain monkey in continuous video frames.

[0137] This embodiment utilizes Figure 3 The algorithm shown is capable of performing action classification, and after performing the action classification, a third recognition result can be obtained based on the third key feature information.

[0138] After the first recognition result, the second recognition result, and the third recognition result are obtained, the first recognition result, the second recognition result, and the third recognition result are taken as a plurality of recognition results.

[0139] When the image processing method of this embodiment is used, it can also be pre-determined whether the target objects in the image to be processed overlap; if the target objects overlap, the image to be processed is segmented. Specifically, the image to be processed can be segmented by using a bounding box regression method to solve the situation where the target objects in the image to be processed overlap, so as to obtain multiple sub-images.

[0140] In addition, before obtaining the image to be processed, it can be pre-determined whether the target object in the current image meets the preset conditions; if the target object in the current image does not meet the preset conditions, an image in which the target object meets the preset conditions is re-obtained, and the image in which the target object meets the preset conditions is used as the image to be processed.

[0141] Specifically, since the image to be processed in this embodiment may include at least two target objects, as one scenario of the preset condition, the number of target objects in the image may be used as the preset condition. Therefore, it can be pre-determined whether the number of target objects in the current image meets the preset number condition. If the number of target objects in the current image does not meet the preset number condition, an image in which the number of target objects meets the preset number condition is obtained again, and the image in which the number of target objects meets the preset number condition is used as the image to be processed. For example, when the number of target objects in the current image is one, the image may not be processed in the manner of steps S201 to S204, until an image including at least two target objects is obtained, and the image is processed in the manner of steps S201 to S204.

[0142] In addition, the definition information of the target object in the image can also be used as a preset condition. When the definition of the target object meets the pixel condition that can be identified, the image whose definition meets the pixel condition that can be identified is used as the image to be processed. Of course, it can be understood that the above preset conditions can also be other conditions related to the target object, which are all within the scope of protection of this application and will not be repeated here.

[0143] At the same time, in order to improve the accuracy of the recognition result, the recognition result for the target object can be displayed to the user, and the user's feedback information on the recognition result can be obtained. For example, when the user's feedback information is that the recognition result is correct, the recognition result can be directly used as the recognition result for the target object in the image to be processed. When the user's feedback information is that the recognition result is wrong, the image to be processed can be re-recognized.

[0144] The embodiment of the present application provides an image processing method, which specifically comprises: obtaining an image to be processed, wherein the image to be processed includes at least two target objects; based on the image to be processed, identifying multiple part areas corresponding to the target objects, obtaining multiple sub-images, wherein the multiple sub-images contain characteristic information of the target objects; associating multiple sub-images corresponding to the same target object with the target object, obtaining multiple sub-images corresponding to the target object; based on the characteristic information of the target object, performing recognition processing on the multiple sub-images corresponding to the target object, and obtaining a recognition result for the target object. When analyzing an image including at least one target object to obtain a recognition result of a target object in the image, the present application associates multiple sub-images corresponding to the same target object in the image with the target object for each target object in the image, and obtains multiple sub-images corresponding to the target object, so that when a target object is recognized, multiple characteristic information of the target object is matched with the target object, so that the subsequent method of identifying the target object is relatively simple. The problem that the existing method of identifying the target object in the image is relatively complicated is solved.

[0145] In the first embodiment described above, an image processing method is provided. Correspondingly, the present application also provides an image processing device. Figure 4 As shown, it is a schematic diagram of an image processing device provided by the second embodiment of the present application. Since the device embodiment is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment. The device embodiment described below is only illustrative.

[0146] This embodiment provides an image processing device, including:

[0147] The image to be processed obtaining unit 401 is used to obtain the image to be processed, wherein the image to be processed includes at least two target objects;

[0148] A sub-image obtaining unit 402 is used to identify multiple part regions corresponding to the target object based on the image to be processed, and obtain multiple sub-images, wherein the multiple sub-images contain feature information of the target object;

[0149] An associating unit 403 is used to associate multiple sub-images corresponding to the same target object with the target object to obtain multiple sub-images corresponding to the target object;

[0150] The target object recognition unit 404 is used to perform recognition processing on the multiple sub-images corresponding to the target object based on the feature information of the target object to obtain a recognition result for the target object.

[0151] Optionally, the target object identification unit is specifically used to:

[0152] Based on the feature information of the target object, respectively perform recognition processing on the multiple sub-images corresponding to the target object to obtain multiple recognition results;

[0153] According to the multiple recognition results, a recognition result for the target object is obtained.

[0154] Optionally, the image to be processed refers to a video frame image;

[0155] The to-be-processed image obtaining unit is specifically used for:

[0156] Sending a request to the video acquisition device to obtain the video file;

[0157] The video file sent by the video acquisition device is received, and a video frame image is obtained according to the video file.

[0158] Optionally, the association unit is specifically used to:

[0159] Get the current target object;

[0160] Based on the feature information of the target object contained in the sub-image, determining a plurality of sub-images matching the current target object;

[0161] Establish an association relationship between the multiple sub-images matching the current target object and the current target object to obtain multiple sub-images corresponding to the target object.

[0162] Optionally, the association unit is specifically used to:

[0163] Based on the feature information of the target object contained in the sub-image, traverse each sub-image of the multiple sub-images to obtain the matching degree information between each sub-image and the current target object;

[0164] According to the matching degree information between each sub-image and the current target object and a preset matching degree threshold condition, a plurality of sub-images matching the current target object are determined.

[0165] Optionally, the association unit is specifically used to:

[0166] Determine whether the matching degree information between the sub-image and the current target object meets the matching degree threshold condition;

[0167] If the sub-image meets the matching degree threshold condition, the sub-image is determined as a sub-image matching the current target object; in the same manner, multiple sub-images matching the current target object are determined.

[0168] Optionally, the sub-image obtaining unit is specifically used to:

[0169] The image to be processed is used as input data of a convolutional neural network and an image region candidate frame generation network to obtain a partition result of the image to be processed; the convolutional neural network and the image region candidate frame generation network are used together as a neural network for partitioning the image to be processed to obtain an image partition result;

[0170] A plurality of sub-images are obtained according to the partitioning result of the image to be processed.

[0171] Optionally, the multiple sub-images include a first sub-image and a second sub-image;

[0172] The target object recognition unit is specifically used to:

[0173] Performing quality assessment on the first sub-image and the second sub-image, and performing first key point positioning and second key point positioning on feature information included in the first sub-image and feature information included in the second sub-image, respectively, to obtain first key feature information and second key feature information that meet preset image quality assessment conditions;

[0174] The first key feature information and the second key feature information that meet the preset image quality assessment condition are identified to obtain a first identification result and the second identification result, and the first identification result and the second identification result are confirmed as the multiple identification results.

[0175] Optionally, the multiple sub-images further include a third sub-image;

[0176] The target object recognition unit is specifically used to:

[0177] Performing third key point positioning on the feature information contained in the third sub-image to obtain third key feature information;

[0178] Classifying the feature information contained in the third sub-image to obtain a category of the feature information of the third sub-image;

[0179] Based on the third key feature information and the feature information category of the third sub-image, a third recognition result is obtained, and the first recognition result, the second recognition result, and the third recognition result are confirmed as the multiple recognition results.

[0180] Optionally, it further includes a first judgment unit and an image segmentation unit; the first judgment unit is used to judge whether the target objects overlap;

[0181] The image segmentation unit is used to segment the image to be processed if the target objects overlap.

[0182] Optionally, it further includes a second judgment unit and an image selection unit; the second judgment unit: judges whether the target object in the current image meets the preset conditions;

[0183] The image selection unit is used to re-obtain an image in which the target object meets the preset condition if the target object in the current image does not meet the preset condition, and use the image in which the target object meets the preset condition as the image to be processed.

[0184] Optionally, it further includes a feedback information obtaining unit and a re-identification unit; the feedback information obtaining unit is used to display the recognition result for the target object and obtain feedback information for the recognition result;

[0185] The re-identification unit is used to determine whether to re-identify the recognition result based on the feedback information.

[0186] The present application also provides an image processing method, such as Figure 5 As shown, it is a flow chart of an image processing method embodiment provided in the third embodiment of the present application. The method comprises the following steps.

[0187] Step S501: obtaining an image to be processed, wherein the processed image includes a target object.

[0188] Step S502: Based on the image to be processed, identify the image to be processed to obtain a plurality of sub-images, wherein the plurality of sub-images contain characteristic information of the target object.

[0189] Step S503: Associating multiple sub-images corresponding to the same target object to obtain multiple sub-images corresponding to the target object.

[0190] Step S504: Based on the feature information of the target object, a plurality of sub-images corresponding to the target object are recognized and processed to obtain a recognition result for the target object.

[0191] It should be noted that the various steps in this embodiment are basically similar to the various steps in the first embodiment described above, and the only difference is that step S502 directly identifies the object to be processed to obtain multiple sub-images, rather than identifying multiple part regions corresponding to the target object to obtain multiple sub-images. Step S503 associates multiple sub-images corresponding to the same target object, rather than associating multiple sub-images corresponding to the same target object with the target object. For the same parts of this embodiment as the first embodiment, please refer to the detailed description of the relevant parts of the first embodiment, which will not be repeated here.

[0192] Based on the third embodiment, the fourth embodiment of the present application provides a traffic image processing method, such as Figure 6As shown, it is a flow chart of a traffic image processing method embodiment provided in the fourth embodiment of the present application. The method comprises the following steps.

[0193] Step S601: obtaining a traffic image to be processed, wherein the traffic image to be processed includes vehicles.

[0194] Step S602: Based on the traffic image to be processed, identify the traffic image to be processed to obtain a plurality of sub-images, wherein the plurality of sub-images contain characteristic information of vehicles.

[0195] Step S603: Associating multiple sub-images corresponding to the same vehicle to obtain multiple sub-images corresponding to the vehicle.

[0196] Step S604: Based on the characteristic information of the vehicle, a plurality of sub-images corresponding to the vehicle are identified and processed to obtain an identification result for the vehicle.

[0197] This embodiment can identify some vehicle information in a traffic scene based on the real-time traffic image obtained by the camera device, for example, it can identify some faulty vehicles or vehicles violating traffic rules.

[0198] Based on the third embodiment, the fifth embodiment of the present application provides an image processing device, which includes: a collection device, a recognition result acquisition device and a display device.

[0199] The acquisition device is used to obtain an image to be processed, wherein the image to be processed includes a target object.

[0200] The recognition result acquisition device is used to identify the image to be processed based on the image to be processed to obtain multiple sub-images, wherein the multiple sub-images contain characteristic information of the target object; associate multiple sub-images corresponding to the same target object to obtain multiple sub-images corresponding to the target object; and perform recognition processing on the multiple sub-images corresponding to the target object based on the characteristic information of the target object to obtain a recognition result for the target object.

[0201] The display device is used to display the recognition result for the target object.

[0202] In the above-mentioned first embodiment, an image processing method is provided. Correspondingly, the sixth embodiment of the present application provides an electronic device corresponding to the method of the first embodiment. Figure 7 As shown, it shows a schematic diagram of the electronic device provided by this embodiment.

[0203] A sixth embodiment of the present application provides an electronic device, including:

[0204] Processor 701;

[0205] The memory 702 is used to store a computer program. The computer program is executed by the processor to perform an image processing method. The method includes the following steps:

[0206] Obtaining an image to be processed, wherein the image to be processed includes at least two target objects;

[0207] Based on the image to be processed, identifying multiple part regions corresponding to the target object, and obtaining multiple sub-images, wherein the multiple sub-images contain feature information of the target object;

[0208] Associating multiple sub-images corresponding to the same target object with the target object to obtain multiple sub-images corresponding to the target object;

[0209] Based on the feature information of the target object, recognition processing is performed on the multiple sub-images corresponding to the target object to obtain a recognition result for the target object.

[0210] In the above-mentioned first embodiment, an image processing method is provided. Correspondingly, the seventh embodiment of the present application provides a computer storage medium corresponding to the method of the first embodiment.

[0211] A seventh embodiment of the present application provides a computer storage medium, wherein the computer storage medium stores a computer program, and the computer program is executed by a processor to perform an image processing method, wherein the method comprises the following steps:

[0212] Obtaining an image to be processed, wherein the image to be processed includes at least two target objects;

[0213] Based on the image to be processed, identifying multiple part regions corresponding to the target object, and obtaining multiple sub-images, wherein the multiple sub-images contain feature information of the target object;

[0214] Associating multiple sub-images corresponding to the same target object with the target object to obtain multiple sub-images corresponding to the target object;

[0215] Based on the feature information of the target object, recognition processing is performed on the multiple sub-images corresponding to the target object to obtain a recognition result for the target object.

[0216] Although the present application is disclosed as above in the form of a preferred embodiment, it is not intended to limit the present application. Any technical personnel in this field may make possible changes and modifications without departing from the spirit and scope of the present application. Therefore, the scope of protection of the present application shall be based on the scope defined by the claims of the present application.

[0217] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory. The memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. Memory is an example of a computer-readable medium.

[0218] 1. Computer readable media includes permanent and non-permanent, removable and non-removable media that can be used to store information by any method or technology. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined in this article, computer readable media does not include non-transitory computer readable storage media (non-transitorycomputer readable storage media), such as modulated data signals and carrier waves.

[0219] 2. Those skilled in the art should understand that the embodiments of the present application can be provided as methods, systems or computer program products. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment or an embodiment combining software and hardware. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

Claims

1. An image processing method, characterized in that: include: Obtaining an image to be processed, wherein the image to be processed includes at least two target objects; Based on the image to be processed, identifying multiple part regions corresponding to the target object, and obtaining multiple sub-images, wherein the multiple sub-images contain feature information of the target object; Associating multiple sub-images corresponding to the same target object with the target object to obtain multiple sub-images corresponding to the target object; Based on the feature information of the target object, respectively perform recognition processing on the multiple sub-images corresponding to the target object to obtain multiple recognition results; According to the multiple recognition results, a recognition result for the target object is obtained.

2. The method according to claim 1, characterized in that: The image to be processed refers to a video frame image; The video frame image is obtained by: Sending a request to the video acquisition device to obtain the video file; The video file sent by the video acquisition device is received, and a video frame image is obtained according to the video file.

3. The method according to claim 1, characterized in that The step of associating a plurality of sub-images corresponding to the same target object with the target object to obtain a plurality of sub-images corresponding to the target object includes: Get the current target object; Based on the feature information of the target object contained in the sub-image, determining a plurality of sub-images matching the current target object; Establish an association relationship between the multiple sub-images matching the current target object and the current target object to obtain multiple sub-images corresponding to the target object.

4. The method according to claim 3, characterized in that The step of determining a plurality of sub-images matching the current target object based on the feature information of the target object contained in the sub-images includes: Based on the feature information of the target object contained in the sub-image, traverse each sub-image of the multiple sub-images to obtain the matching degree information between each sub-image and the current target object; According to the matching degree information between each sub-image and the current target object and a preset matching degree threshold condition, a plurality of sub-images matching the current target object are determined.

5. The method according to claim 4, characterized in that The determining, according to the matching degree information between each sub-image and the current target object and the matching degree threshold condition, a plurality of sub-images matching the current target object comprises: Determine whether the matching degree information between the sub-image and the current target object meets the matching degree threshold condition; If the sub-image meets the matching degree threshold condition, the sub-image is determined as a sub-image matching the current target object; in the same manner, multiple sub-images matching the current target object are determined.

6. The method according to claim 1, characterized in that The step of identifying a plurality of part regions corresponding to the target object based on the image to be processed to obtain a plurality of sub-images includes: The image to be processed is used as input data of a convolutional neural network and an image region candidate frame generation network to obtain a partition result of the image to be processed; the convolutional neural network and the image region candidate frame generation network are used together as a neural network for partitioning the image to be processed to obtain an image partition result; A plurality of sub-images are obtained according to the partitioning result of the image to be processed.

7. The method according to claim 1, characterized in that The multiple sub-images include a first sub-image and a second sub-image; based on the feature information of the target object, the multiple sub-images corresponding to the target object are respectively identified and processed to obtain multiple identification results, including: Performing quality assessment on the first sub-image and the second sub-image, and performing first key point positioning and second key point positioning on feature information included in the first sub-image and feature information included in the second sub-image, respectively, to obtain first key feature information and second key feature information that meet preset image quality assessment conditions; The first key feature information and the second key feature information that meet the preset image quality assessment condition are identified to obtain a first identification result and a second identification result, and the first identification result and the second identification result are confirmed as the multiple identification results.

8. The method according to claim 7, characterized in that The multiple sub-images also include a third sub-image; based on the feature information of the target object, the multiple sub-images corresponding to the target object are respectively identified and processed to obtain multiple identification results, including: Performing third key point positioning on the feature information contained in the third sub-image to obtain third key feature information; Classifying the feature information contained in the third sub-image to obtain a category of the feature information of the third sub-image; Based on the third key feature information and the feature information category of the third sub-image, a third recognition result is obtained, and the first recognition result, the second recognition result, and the third recognition result are confirmed as the multiple recognition results.

9. The method according to claim 1, characterized in that: Also includes: Determining whether the target objects overlap; If the target objects overlap, segmentation processing is performed on the image to be processed.

10. The method according to claim 1, characterized in that Also includes: Determine whether the target object in the current image meets the preset conditions; If the target object in the current image does not meet the preset condition, an image in which the target object meets the preset condition is obtained again, and the image in which the target object meets the preset condition is used as the image to be processed.

11. The method according to claim 1, characterized in that: Also includes: Displaying the recognition result for the target object and obtaining feedback information for the recognition result; Based on the feedback information, it is determined whether to re-identify the recognition result.

12. An image processing device, characterized in that: include: A unit for obtaining an image to be processed, used for obtaining an image to be processed, wherein the image to be processed includes at least two target objects; A sub-image obtaining unit, configured to identify a plurality of part regions corresponding to the target object based on the image to be processed, and obtain a plurality of sub-images, wherein the plurality of sub-images contain feature information of the target object; an associating unit, used to associate multiple sub-images corresponding to the same target object with the target object to obtain multiple sub-images corresponding to the target object; The target object recognition unit is used to perform recognition processing on the multiple sub-images corresponding to the target object based on the feature information of the target object to obtain multiple recognition results; and obtain a recognition result for the target object based on the multiple recognition results.

13. An image processing method, characterized in that: include: Obtaining an image to be processed, wherein the image to be processed includes a target object; Based on the image to be processed, identifying the image to be processed, and obtaining a plurality of sub-images, wherein the plurality of sub-images contain feature information of a target object; Associating multiple sub-images corresponding to the same target object to obtain multiple sub-images corresponding to the target object; Based on the feature information of the target object, respectively perform recognition processing on the multiple sub-images corresponding to the target object to obtain multiple recognition results; According to the multiple recognition results, a recognition result for the target object is obtained.

14. A traffic image processing method, characterized in that: include: Obtaining a traffic image to be processed, wherein the traffic image to be processed includes vehicles; Based on the traffic image to be processed, identifying the traffic image to be processed, and obtaining a plurality of sub-images, wherein the plurality of sub-images contain characteristic information of vehicles; Associating multiple sub-images corresponding to the same vehicle to obtain multiple sub-images corresponding to the vehicle; Based on the characteristic information of the vehicle, respectively perform recognition processing on the multiple sub-images corresponding to the vehicle to obtain multiple recognition results; According to the multiple recognition results, a recognition result for the vehicle is obtained.

15. An image processing device, characterized in that: include: Collection device, recognition result acquisition device; The acquisition device is used to obtain an image to be processed, wherein the image to be processed includes a target object; The recognition result acquisition device is used to recognize the image to be processed based on the image to be processed, and obtain multiple sub-images, wherein the multiple sub-images contain feature information of the target object; Associating multiple sub-images corresponding to the same target object to obtain multiple sub-images corresponding to the target object; Based on the feature information of the target object, respectively perform recognition processing on the multiple sub-images corresponding to the target object to obtain multiple recognition results; According to the multiple recognition results, a recognition result for the target object is obtained.

16. The image processing device according to claim 15, characterized in that It also includes a display device; the display device is used to display the recognition result for the target object.

17. An electronic device, characterized in that: include: processor; The memory is used to store a computer program, the computer program is executed by the processor to perform an image processing method, the method comprising the following steps: Obtaining an image to be processed, wherein the image to be processed includes at least two target objects; Based on the image to be processed, identifying multiple part regions corresponding to the target object, and obtaining multiple sub-images, wherein the multiple sub-images contain feature information of the target object; Associating multiple sub-images corresponding to the same target object with the target object to obtain multiple sub-images corresponding to the target object; Based on the feature information of the target object, respectively perform recognition processing on the multiple sub-images corresponding to the target object to obtain multiple recognition results; According to the multiple recognition results, a recognition result for the target object is obtained.

18. A computer storage medium, characterized in that: The computer storage medium stores a computer program, which is executed by the processor to perform an image processing method, the method comprising the following steps: Obtaining an image to be processed, wherein the image to be processed includes at least two target objects; Based on the image to be processed, identifying multiple part regions corresponding to the target object, and obtaining multiple sub-images, wherein the multiple sub-images contain feature information of the target object; Associating multiple sub-images corresponding to the same target object with the target object to obtain multiple sub-images corresponding to the target object; Based on the feature information of the target object, respectively perform recognition processing on the multiple sub-images corresponding to the target object to obtain multiple recognition results; According to the multiple recognition results, a recognition result for the target object is obtained.

Images