Image processing method and device, electronic equipment and computer storage medium

By enhancing data and reducing resolution of the training images of the autonomous driving lane line recognition model, the problem of lane line image breaking during feature extraction is solved, the recognition accuracy is improved, and the continuous recognition of lane line is ensured.

CN118918560APending Publication Date: 2024-11-08BEIJING ZHIXINGZHE TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202310505162.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2023-05-06
Publication Date
2024-11-08

AI Technical Summary

Technical Problem

In autonomous driving, the lane line recognition model has a lower recognition accuracy because the thinner lane line image in the training image breaks during feature extraction.

Method used

By performing data enhancement processing and resolution reduction processing on the original image, the target image is obtained and used to train and recognize the model, avoiding the fracture problem caused by too small lane line image size.

Benefits of technology

The recognition accuracy of the lane line recognition model is improved, ensuring the continuity and accuracy of the lane line during the recognition process.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118918560A_ABST
    Figure CN118918560A_ABST
Patent Text Reader

Abstract

The embodiment of the invention provides an image processing method and device, electronic equipment and a computer storage medium. The method comprises the steps that an original image including a to-be-recognized object image is acquired; performing data enhancement processing on a to-be-recognized object in the original image to obtain an enhanced image; and performing resolution reduction processing on the enhanced image to obtain a target image, the target image being used for training a recognition model, and the recognition model being used for recognizing an object to be recognized from an input image. The target image obtained by using the image processing method provided by the invention is subjected to model training, and the trained recognition model has high recognition accuracy on the to-be-recognized object.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

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

[0002] In autonomous driving, vision-based detection plays an important role. Through the mutual complementation of visual detection results and other sensors, the surrounding environment information can be obtained in real time, thus providing necessary information for autonomous driving planning and decision-making. Lane lines are important landmarks in autonomous driving scenarios. The recognition of lane lines is closely related to the subsequent vehicle positioning and autonomous driving planning and decision-making. Therefore, a separate lane line recognition model is needed to recognize lane lines.

[0003] Currently, a lane line recognition model is trained by acquiring training images labeled with lane line images.

[0004] However, when the lane line recognition model is trained through training images, the thinner lane line images in the training images will be broken due to feature extraction, which will cause the trained lane line recognition model to have broken lane lines when recognizing lane lines. Therefore, the recognition accuracy of the trained lane line recognition model is low. Summary of the invention

[0005] In view of this, embodiments of the present application provide an image processing method, device, electronic device, and computer storage medium to at least partially solve the above problems.

[0006] According to a first aspect of an embodiment of the present application, there is provided an image processing method, comprising: acquiring an original image including an image of an object to be identified; performing data enhancement processing on the object to be identified in the original image to obtain an enhanced image; performing resolution reduction processing on the enhanced image to obtain a target image, wherein the target image is used to train a recognition model, and the recognition model is used to identify the object to be identified from an input image.

[0007] In a possible implementation, performing data enhancement processing on the object to be identified in the original image to obtain an enhanced image includes: performing maximum pooling processing and / or dilation processing on the image of the object to be identified in the original image to obtain the enhanced image that matches the contour of the image of the object to be identified in the original image.

[0008] In a possible implementation, the performing resolution reduction processing on the enhanced image to obtain the target image includes: performing nearest neighbor interpolation processing on the enhanced image to obtain the target image.

[0009] In a possible implementation manner, the object to be identified includes at least one of a lane line, a lane edge, a road guardrail, a street light pole, and a road sign pole.

[0010] According to the second aspect of the embodiments of the present application, a recognition model training method is provided, comprising: obtaining a training set comprising multiple training samples, wherein the training samples include the target image obtained by the method described in the first aspect of the embodiments of the present application; training a model to be trained by using the training set to obtain a recognition model for identifying an object to be identified from an input image.

[0011] According to a third aspect of an embodiment of the present application, an image processing device is provided, comprising: a first acquisition module, used to acquire an original image including an image of an object to be identified; an extraction module, used to perform data enhancement processing on the object to be identified in the original image to obtain an enhanced image; and a processing module, used to perform resolution reduction processing on the enhanced image to obtain a target image, wherein the target image is used to train a recognition model, and the recognition model is used to identify the object to be identified from an input image.

[0012] In a possible implementation, the extraction module is used to perform maximum pooling processing and / or dilation processing on the image of the object to be identified in the original image to obtain the enhanced image that matches the contour of the image of the object to be identified in the original image.

[0013] According to the fourth aspect of the embodiments of the present application, a recognition model training device is provided, comprising: a second acquisition module, used to acquire a training set including multiple training samples, wherein the training samples include the target image obtained by the method described in the first aspect of the embodiments of the present application; a training module, used to train a model to be trained by using the training set to obtain a recognition model for identifying an object to be identified from an input image.

[0014] According to the fifth aspect of the embodiments of the present application, there is provided an electronic device, comprising: a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other through the communication bus; the memory is used to store at least one executable instruction, and the executable instruction enables the processor to perform an operation corresponding to the method described in the first aspect or the second aspect.

[0015] According to a sixth aspect of an embodiment of the present application, a computer storage medium is provided, on which a computer program is stored. When the program is executed by a processor, the method described in the first aspect or the second aspect is implemented.

[0016] According to the image processing scheme provided in the embodiment of the present application, data enhancement processing is performed on the image of the object to be identified in the original image to obtain an enhanced image, and the resolution of the enhanced image is reduced to obtain a target image, thereby increasing the size of the image of the object to be identified, thereby avoiding the situation in which the image of the object to be identified is broken due to the small size of the image of the object to be identified during the training of the recognition model, thereby avoiding the situation in which the trained recognition model identifies the object to be identified and the identified object is broken, so the recognition model trained by the target image has a higher recognition accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] 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 embodiments of the present application. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0018] Figure 1 is a flow chart of an image processing method provided by an embodiment of the present application;

[0019] Figure 2 is a flow chart of a recognition model training method provided in an embodiment of the present application;

[0020] Figure 3 This is a recognition effect diagram of a lane line by a recognition model provided in an embodiment of the present application;

[0021] Figure 4 This is a recognition effect diagram of lane lines by another recognition model provided in an embodiment of the present application;

[0022] Figure 5 is a schematic diagram of an image processing device provided in an embodiment of the present application;

[0023] Figure 6 is a schematic diagram of a recognition model training device provided in an embodiment of the present application;

[0024] Figure 7 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0025] In order to enable those skilled in the art to better understand the technical solutions in the embodiments of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments in the embodiments of the present application should fall within the scope of protection of the embodiments of the present application.

[0026] The terms used in this application are for the purpose of describing specific embodiments only and are not intended to limit this application. The singular forms of "a", "said" and "the" used in this application and the appended claims are also intended to include plural forms unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used herein refers to and includes any or all possible combinations of one or more associated listed items.

[0027] It should be understood that although the terms first, second, third, etc. may be used in the present application to describe various information, these information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of the present application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0028] As mentioned above, vision-based detection plays an important role in autonomous driving. Through the mutual complementation of visual detection results and other sensors, the surrounding environmental information can be obtained in real time. The environmental information includes pedestrians, car drivable areas, obstacles, lane lines, etc., which can provide necessary information for autonomous driving planning and decision-making. Lane lines are important landmarks in autonomous driving scenarios. The recognition of lane lines is closely related to the subsequent vehicle positioning and autonomous driving planning and decision-making. Therefore, a separate lane line recognition model is needed to recognize lane lines.

[0029] At present, a lane line recognition model is trained by obtaining training images with lane line images labeled thereon. However, when the lane line recognition model is trained by using training images, thinner lane line images in the training images may be broken due to feature extraction. For example, since the lane lines farther away in the training images are thinner, the lane lines farther away may be broken during the convolution process, making it impossible for the lane line recognition model to recognize the complete lane lines. This in turn causes the trained lane line recognition model to have broken lane lines when recognizing lane lines. Therefore, the recognition accuracy of the trained lane line recognition model is low.

[0030] In an embodiment of the present application, data enhancement processing is performed on the image of the object to be identified in the original image to obtain an enhanced image, and the enhanced image is subjected to resolution reduction processing to obtain a target image, thereby increasing the size of the image of the object to be identified, thereby avoiding the situation in which the image of the object to be identified is broken due to the small size of the image of the object to be identified during the training of the recognition model, thereby avoiding the situation in which the trained recognition model identifies the object to be identified and the identified object is broken, so the recognition model trained by the target image has a higher recognition accuracy.

[0031] The specific implementation of the embodiment of the present application is further explained below in conjunction with the accompanying drawings of the embodiment of the present application.

[0032] Figure 1 is a flow chart of an image processing method provided in an embodiment of the present application, such as Figure 1 As shown, the method includes the following steps 101 to 103:

[0033] Step 101: Acquire an original image including an image of an object to be identified.

[0034] The original image is obtained. The original image may be a manually annotated image including an image of an object to be identified. The object to be identified may be any object, such as lane lines, trees, etc.

[0035] Step 102: Perform data enhancement processing on the image of the object to be identified in the original image to obtain an enhanced image.

[0036] Data enhancement is performed on the image of the object to be identified in the original image, and the image of the object to be identified after the data enhancement is determined as the enhanced image. Through the data enhancement process, the image of the object to be identified in the enhanced image obtained after the data enhancement becomes coarser and clearer.

[0037] Step 103: Perform resolution reduction processing on the enhanced image to obtain a target image.

[0038] The target image is used to train the recognition model, and the recognition model is used to recognize the object to be recognized from the input image.

[0039] The enhanced image is processed by reducing the resolution so that the resolution of the target image is smaller, so that the recognition model to be trained can be trained with the target image.

[0040] In an embodiment of the present application, data enhancement processing is performed on the image of the object to be identified in the original image to obtain an enhanced image, and the enhanced image is subjected to resolution reduction processing to obtain a target image, thereby increasing the size of the image of the object to be identified, thereby avoiding the situation in which the image of the object to be identified is broken due to the small size of the image of the object to be identified during the training of the recognition model, thereby avoiding the situation in which the trained recognition model identifies the object to be identified and the identified object is broken, so the recognition model trained by the target image has a higher recognition accuracy.

[0041] Optionally, the original image may be first processed by reducing the resolution, and then the image of the object to be identified in the original image after the resolution is reduced may be data enhanced to obtain an enhanced image with a lower resolution. This may increase the size of the image of the object to be identified, thereby avoiding the situation in which the image of the object to be identified is broken due to its small size during the training of the recognition model. This may also avoid the situation in which the trained recognition model identifies the object to be identified with broken images. Therefore, the recognition model trained using the target image has a higher recognition accuracy.

[0042] In one possible implementation, when performing data enhancement processing on the object to be identified in the original image to obtain an enhanced image, maximum pooling processing and / or dilation processing can be performed on the image of the object to be identified in the original image to obtain an enhanced image that matches the contour of the image of the object to be identified in the original image.

[0043] The object to be identified in the original image is subjected to maximum pooling and / or dilation processing. The size of the pooling kernel of the maximum pooling processing can be selected according to the resolution size of the image of the object to be identified. A larger pooling kernel can be selected for an image with a higher resolution, and a smaller pooling kernel can be selected for an image with a lower resolution. For example, a 4×4 pooling kernel is used for an image with a higher resolution, and a 2×2 pooling kernel is used for an image with a lower resolution, and so on.

[0044] When the object to be identified in the original image is expanded, the expanded pixels can be selected according to the resolution of the image of the object to be identified. For an image with a higher resolution, more pixels are expanded, and for an image with a lower resolution, fewer pixels are expanded. For example, an image with a higher resolution is expanded by 3 pixels, and an image with a lower resolution is expanded by 1 pixel, and so on.

[0045] It should be noted that the object to be identified can be subjected to maximum pooling processing or dilation processing, or both. In this case, the order of maximum pooling processing and dilation processing is not fixed, that is, maximum pooling processing can be performed first, followed by dilation processing, or dilation processing can be performed first and then maximum pooling processing.

[0046] The enhanced image matches the contour of the object to be identified in the original image. For example, the replacement image of the tree image is similar to the tree image in the original image in shape and contour, both are rectangles (two-dimensional cross-sections of cylinders), or the replacement image of the lane line is linear to the lane line image in the original image.

[0047] In an embodiment of the present application, an enhanced image is obtained through maximum pooling processing and / or dilation processing. The image of the object to be identified can be coarsened through maximum pooling processing and / or dilation processing, so that the enhanced image can accurately represent the shape and other features of the object to be identified, thereby making the recognition accuracy of the trained recognition model higher.

[0048] In a possible implementation, when the enhanced image is subjected to resolution reduction processing to obtain the target image, the enhanced image may be subjected to nearest neighbor interpolation processing to obtain the target image.

[0049] The enhanced image is interpolated by the nearest neighbor to reduce the resolution of the enhanced image.

[0050] In an embodiment of the present application, the target image is obtained by performing nearest neighbor interpolation on the enhanced image, so that the resolution of the enhanced image can be reduced, which shortens the time required for training the recognition model. Therefore, the efficiency of training the recognition model through the target image is higher, thereby improving the efficiency of training the model.

[0051] In a possible implementation manner, the object to be identified includes at least one of a lane line, a lane edge, a road guardrail, a street light pole, and a road sign pole.

[0052] In the embodiment of the present application, at least one of the lane lines, lane edges, road guardrails, street lamp poles, and road sign poles to be identified can be enhanced through data enhancement so that the images of objects such as lane lines, lane edges, road guardrails, street lamp poles, and road sign poles that are farther away will not be broken during the model training process. Therefore, the trained recognition model has a high recognition accuracy in identifying lane lines, lane edges, road guardrails, street lamp poles, or road sign poles.

[0053] Figure 2 is a flow chart of a recognition model training method provided in an embodiment of the present application, such as Figure 2 As shown, the method includes the following steps 201 to 202:

[0054] Step 201: Obtain a training set including multiple training samples.

[0055] The training sample includes a target image obtained by any of the above-mentioned image processing method embodiments.

[0056] Step 202: Train the model to be trained using the training set to obtain a recognition model for recognizing the object to be recognized from the input image.

[0057] The model to be trained is trained using the target image to obtain a recognition model.

[0058] For example, Figure 3 The figure shows the recognition effect of the recognition model trained by the original image on the lane line. Figure 4 The figure shows the recognition effect of the recognition model trained by the target image on the lane line. Figure 3 and Figure 4 As shown, in Figure 3 The boxed area shows that the recognition model trained with the original image breaks down when recognizing lane lines farther away, while Figure 4 At the same position, when the recognition model trained by the target image recognizes the lane lines farther away, the lane lines are continuous, so the recognition model trained by the target image has a higher recognition accuracy.

[0059] In an embodiment of the present application, a recognition model is obtained by training a model to be trained with a target image. Since the size of the image of the object to be identified increases after image processing, the situation in which the image of the object to be identified is broken due to its small size when extracting features from the image of the object to be identified during the training of the recognition model is avoided. This can prevent the trained recognition model from identifying the object to be identified with broken parts. Therefore, the recognition accuracy of the trained recognition model is higher.

[0060] Figure 5 is a schematic diagram of an image processing device provided in an embodiment of the present application, such as Figure 5 As shown, the image processing device 500 includes:

[0061] The first acquisition module 501 is used to acquire an original image including an image of an object to be identified.

[0062] The extraction module 502 is used to perform data enhancement processing on the object to be identified in the original image to obtain an enhanced image.

[0063] The processing module 503 is used to perform resolution reduction processing on the enhanced image to obtain a target image, wherein the target image is used to train a recognition model, and the recognition model is used to recognize an object to be recognized from an input image.

[0064] In the embodiment of the present application, the first acquisition module 501 can be used to execute step 101 in the above method embodiment, the extraction module 502 can be used to execute step 102 in the above method embodiment, and the processing module 503 can be used to execute step 103 in the above method embodiment.

[0065] In a possible implementation, the extraction module 502 may perform maximum pooling processing and / or dilation processing on the image of the object to be identified in the original image to obtain an enhanced image that matches the contour of the image of the object to be identified in the original image.

[0066] In a possible implementation, the processing module 503 may perform nearest neighbor interpolation processing on the enhanced image to obtain a target image.

[0067] In a possible implementation manner, the object to be identified includes at least one of a lane line, a lane edge, a road guardrail, a street light pole, and a road sign pole.

[0068] It should be noted that, since the information interaction, execution process, etc. between the modules in the above-mentioned image processing device are based on the same concept as the above-mentioned image processing method embodiment, the specific contents can be found in the description in the above-mentioned image processing method embodiment and will not be repeated here.

[0069] Figure 6 is a schematic diagram of a recognition model training device provided in an embodiment of the present application, such as Figure 6 As shown, the recognition model training device 600 includes:

[0070] The second acquisition module 601 is used to acquire a training set including a plurality of training samples, wherein the training samples include a target image obtained by any of the above-mentioned embodiments.

[0071] The training module 602 is used to train the to-be-trained model using a training set to obtain a recognition model for recognizing the to-be-recognized object from an input image.

[0072] In the embodiment of the present application, the second acquisition module 601 may be used to execute step 201 in the above method embodiment, and the training module 602 may be used to execute step 202 in the above method embodiment.

[0073] It should be noted that, since the information interaction, execution process, etc. between the modules in the above-mentioned recognition model training device are based on the same concept as the above-mentioned recognition model training method embodiment, the specific contents can be found in the description in the above-mentioned recognition model training method embodiment and will not be repeated here.

[0074] Reference Figure 7 , shows a schematic diagram of the structure of an electronic device according to an embodiment of the present application. The specific embodiment of the present application does not limit the specific implementation of the electronic device.

[0075] like Figure 7As shown, the electronic device may include: a processor (processor) 702 , a communication interface (Communications Interface) 704 , a memory (memory) 706 , and a communication bus 708 .

[0076] in:

[0077] The processor 702 , the communication interface 704 , and the memory 706 communicate with each other via a communication bus 708 .

[0078] The communication interface 704 is used to communicate with other electronic devices or servers.

[0079] The processor 702 is used to execute the program 710, and specifically can execute the relevant steps in the above-mentioned image processing method or recognition model training method embodiment.

[0080] Specifically, the program 710 may include program codes, which include computer operation instructions.

[0081] The processor 702 may be a central processing unit (CPU), a graphics processing unit (GPU), an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application. The one or more processors included in the smart device may be processors of the same type, such as one or more CPUs; one or more GPUs; or processors of different types, such as one or more CPUs, one or more GPUs, and one or more ASICs.

[0082] The memory 706 is used to store the program 710. The memory 706 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

[0083] The program 710 can be specifically used to enable the processor 702 to execute the image processing method or the recognition model training method in any of the aforementioned embodiments.

[0084] The specific implementation of each step in program 710 can refer to the corresponding description in the corresponding steps and units in any of the above-mentioned image processing methods or recognition model training method embodiments, which will not be repeated here. Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working process of the above-described devices and modules can refer to the corresponding process description in the above-mentioned method embodiments, which will not be repeated here.

[0085] In an embodiment of the present application, data enhancement processing is performed on the image of the object to be identified in the original image to obtain a replacement image, and the image of the object to be identified in the original image is pixel-expanded according to the replacement image to obtain a target image, thereby increasing the size of the image of the object to be identified, thereby avoiding the situation where the image of the object to be identified is broken due to the small size of the image of the object to be identified when feature extraction is performed on the image of the object to be identified during the training of the recognition model, thereby avoiding the situation where the object to be identified is broken when the trained recognition model identifies the object to be identified, and therefore the recognition accuracy of the recognition model trained by the target image is higher.

[0086] An embodiment of the present application further provides a computer storage medium on which a computer program is stored. When the program is executed by a processor, the method described in any one of the above-mentioned method embodiments is implemented.

[0087] An embodiment of the present application also provides a computer program product, including computer instructions, which instruct a computing device to execute operations corresponding to any one of the above-mentioned multiple method embodiments.

[0088] It should be pointed out that, according to the needs of implementation, the various components / steps described in the embodiments of the present application can be split into more components / steps, or two or more components / steps or partial operations of components / steps can be combined into new components / steps to achieve the purpose of the embodiments of the present application.

[0089] The above-mentioned method according to the embodiment of the present application can be implemented in hardware, firmware, or implemented as software or computer code that can be stored in a recording medium (such as a CD ROM, RAM, floppy disk, hard disk or magneto-optical disk), or implemented as computer code originally stored in a remote recording medium or a non-temporary machine-readable medium downloaded through a network and to be stored in a local recording medium, so that the method described herein can be stored in such software processing on a recording medium using a general-purpose computer, a dedicated processor or programmable or dedicated hardware (such as an ASIC or FPGA). It can be understood that a computer, a processor, a microprocessor controller or programmable hardware includes a storage component (e.g., RAM, ROM, flash memory, etc.) that can store or receive software or computer code, and when the software or computer code is accessed and executed by a computer, a processor or hardware, the image processing method described herein is implemented. In addition, when a general-purpose computer accesses the code for implementing the image processing method shown herein, the execution of the code converts the general-purpose computer into a dedicated computer for executing the image processing method shown herein.

[0090] Those of ordinary skill in the art will appreciate that the units and method steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the embodiments of the present application.

[0091] The above implementation methods are only used to illustrate the embodiments of the present application, and are not limitations on the embodiments of the present application. Ordinary technicians in the relevant technical field can make various changes and modifications without departing from the spirit and scope of the embodiments of the present application. Therefore, all equivalent technical solutions also belong to the scope of the embodiments of the present application. The scope of patent protection of the embodiments of the present application should be limited by the claims.

Claims

1. An image processing method, characterized in that: include: Acquire an original image including an image of an object to be identified; Performing data enhancement processing on the object to be identified in the original image to obtain an enhanced image; The enhanced image is subjected to resolution reduction processing to obtain a target image, wherein the target image is used to train a recognition model, and the recognition model is used to recognize an object to be recognized from an input image.

2. The method according to claim 1, characterized in that: The step of performing data enhancement processing on the object to be identified in the original image to obtain an enhanced image includes: The image of the object to be identified in the original image is subjected to maximum pooling processing and / or dilation processing to obtain the enhanced image that matches the contour of the image of the object to be identified in the original image.

3. The method according to claim 2, characterized in that The step of performing resolution reduction processing on the enhanced image to obtain a target image comprises: Perform nearest neighbor interpolation processing on the enhanced image to obtain the target image.

4. The method according to any one of claims 1 to 3, characterized in that: The object to be identified includes at least one of a lane line, a lane edge, a road guardrail, a street light pole, and a road sign pole.

5. A recognition model training method, characterized in that: include: Acquire a training set including a plurality of training samples, wherein the training samples include the target image obtained by the method according to any one of claims 1 to 4; The model to be trained is trained by using the training set to obtain a recognition model for recognizing the object to be recognized from the input image.

6. An image processing device, characterized in that: include: A first acquisition module, used to acquire an original image including an image of an object to be identified; An extraction module is used to perform data enhancement processing on the object to be identified in the original image to obtain an enhanced image; The processing module is used to perform resolution reduction processing on the enhanced image to obtain a target image, wherein the target image is used to train a recognition model, and the recognition model is used to recognize an object to be recognized from an input image.

7. The device according to claim 6, characterized in that The extraction module is used to perform maximum pooling processing and / or dilation processing on the image of the object to be identified in the original image to obtain the enhanced image that matches the contour of the image of the object to be identified in the original image.

8. A recognition model training device, characterized in that: include: A second acquisition module, configured to acquire a training set including a plurality of training samples, wherein the training samples include the target image obtained by the method according to any one of claims 1 to 4; The training module is used to train the to-be-trained model using the training set to obtain a recognition model for recognizing the to-be-recognized object from the input image.

9. An electronic device, characterized in that: include: A processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other via the communication bus; The memory is used to store at least one executable instruction, and the executable instruction enables the processor to perform operations corresponding to the image processing method as described in any one of claims 1 to 4 or the recognition model training method as described in claim 5.

10. A computer storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the image processing method as described in any one of claims 1 to 4 or the recognition model training method as described in claim 5 is implemented.