Image recognition method and device based on deep learning and storage medium

Abstract

The invention relates to the technical field of image processing, in particular to an image recognition method and device based on deep learning, computer equipment, a storage medium and a computer program product. The method comprises the following steps: acquiring an initial feature recognition network model, and screening out a target processing unit meeting a preset removal index from processing units of the initial feature recognition network model; removing the target processing unit from the initial feature recognition network model to obtain a target feature recognition network model; obtaining a to-be-processed image, and processing the to-be-processed image based on a preset preprocessing process to obtain a target image; inputting the target image into a target feature recognition network model for processing to obtain a recognition result; and determining feature information of a feature unit in the to-be-processed image according to the identification result. By adopting the method, the image recognition efficiency can be improved.

Description

Technical Field

[0001] The present application relates to the field of image processing technology, and in particular to an image recognition method, apparatus, computer equipment, storage medium and computer program product based on deep learning. Background Art

[0002] Image recognition technology refers to the process of automatically identifying and classifying objects, scenes or features in an image using computer vision and pattern recognition technology. By analyzing and processing images, image recognition technology can achieve tasks such as target detection, object recognition, face recognition, image classification, and image segmentation, thereby realizing the understanding and recognition of image content. The main purposes of image recognition include: Automated processing: Image recognition technology can realize the automated processing and analysis of large amounts of image data, thereby improving work efficiency and reducing labor costs. For example, in industrial production, image recognition technology can be used to perform quality inspection and classification of products. Intelligent monitoring: Image recognition technology can be applied to intelligent monitoring systems to achieve real-time monitoring and identification of objects and activities in the environment. For example, in intelligent transportation systems, image recognition technology can be used to monitor and identify vehicles and pedestrians. Security: Image recognition technology can be applied to the field of security to achieve identification and monitoring of personnel identity, behavior, and activities. For example, face recognition technology can be used to authenticate the identity of people entering and leaving. Intelligent assistance: Image recognition technology can be applied to intelligent assistance systems to provide users with intelligent services and support. For example, image recognition technology can be used to identify and recommend commodities. In general, image recognition technology can help people better understand and use image information, realize automatic recognition and analysis of image content, and thus be applied to various fields to improve work efficiency, enhance security, provide intelligent services, etc. Therefore, image recognition technology has broad application prospects and important significance in modern society.

[0003] In traditional technology, deep learning-based image recognition methods refer to methods that use deep learning technology to perform tasks such as image classification, target detection, and image segmentation. Deep learning is a machine learning technology that builds a multi-layer neural network to learn the complex feature representation of data, thereby achieving efficient processing and analysis of data such as images, text, and speech. In image recognition tasks, deep learning methods usually use convolutional neural networks (CNN) as the basic model. CNN is a neural network structure specially designed to process image data. It extracts image features and classifies them through components such as convolutional layers, pooling layers, and fully connected layers.

[0004] However, the current image recognition methods have the following technical problems:

[0005] Using a large model to process and identify images can easily lead to redundant calculations due to the full functionality of the model, and the efficiency of image recognition processing is low. Summary of the invention

[0006] Based on this, it is necessary to provide an image recognition method, device, computer equipment, computer-readable storage medium and computer program product based on deep learning that can improve the efficiency of image recognition in order to address the above technical problems.

[0007] In a first aspect, the present application provides an image recognition method based on deep learning. The method comprises:

[0008] Acquire an image to be processed, and process the image to be processed based on a preset preprocessing process to obtain a target image;

[0009] Inputting the target image into a target feature recognition network model for processing to obtain a recognition result, wherein the processing unit of the target feature recognition network model is less than that of the initial feature recognition network model;

[0010] Determine feature information of feature units in the image to be processed according to the recognition result.

[0011] In one embodiment, the step of inputting the target image into a target feature recognition network model for processing and obtaining a recognition result includes:

[0012] Acquire an initial feature recognition network model, and select target processing units that meet preset removal indicators from the processing units of the initial feature recognition network model;

[0013] The target processing unit is removed from the initial feature recognition network model to obtain the target feature recognition network model.

[0014] In one embodiment, the obtaining of the initial feature recognition network model and selecting the target processing unit that meets the preset removal index from the processing units of the initial feature recognition network model comprises:

[0015] Acquire a historical sample set, wherein the historical sample set includes paired historical image samples and historical image feature information;

[0016] Traversing the processing unit of the initial feature recognition network model without the processing unit, and verifying the accuracy of the result of the initial feature recognition network model after the processing unit is removed through the historical sample set;

[0017] The target processing unit is determined according to the verification result of the result accuracy.

[0018] In one embodiment, the method further comprises:

[0019] In response to different feature recognition requirements, obtaining the historical sample set corresponding to the feature recognition requirements;

[0020] The target feature recognition network model is reacquired based on the updated historical sample set.

[0021] In one of the embodiments, the target processing unit includes a combination of one or more of a model layer, a model module, a model channel, and a model calculation parameter.

[0022] In one embodiment, the preprocessing process includes one or more combinations of denoising processing, image enhancement processing, resizing processing, color space conversion processing, edge detection processing, and image smoothing processing.

[0023] In a second aspect, the present application also provides an image recognition device based on deep learning. The device comprises:

[0024] A preprocessing module is used to obtain an image to be processed, and process the image to be processed based on a preset preprocessing process to obtain a target image;

[0025] A model recognition module, used for inputting the target image into a target feature recognition network model for processing to obtain a recognition result, wherein the processing unit of the target feature recognition network model is less than that of the initial feature recognition network model;

[0026] The feature information module is used to determine the feature information of the feature unit in the image to be processed according to the recognition result.

[0027] In one embodiment, the model identification module includes:

[0028] A processing unit screening module, used to obtain an initial feature recognition network model, and screen out target processing units that meet preset removal indicators from the processing units of the initial feature recognition network model;

[0029] The model pruning module is used to remove the target processing unit from the initial feature recognition network model to obtain the target feature recognition network model.

[0030] In one embodiment, the processing unit screening module includes:

[0031] A historical sample set module, used to obtain a historical sample set, wherein the historical sample set includes paired historical image samples and historical image feature information;

[0032] An accuracy verification module, used for traversing the processing unit of the initial feature recognition network model after removing the processing unit, and verifying the accuracy of the result of the initial feature recognition network model after removing the processing unit through the historical sample set;

[0033] The target processing unit module is used to determine the target processing unit according to the verification result of the result accuracy.

[0034] In one embodiment, the device further comprises:

[0035] An identification requirement module, used for obtaining the historical sample set corresponding to the feature identification requirement in response to different feature identification requirements;

[0036] A model updating module is used to reacquire the target feature recognition network model based on the updated historical sample set.

[0037] In one of the embodiments, the target processing unit includes a combination of one or more of a model layer, a model module, a model channel, and a model calculation parameter.

[0038] In one embodiment, the preprocessing process includes one or more combinations of denoising processing, image enhancement processing, resizing processing, color space conversion processing, edge detection processing, and image smoothing processing.

[0039] In a third aspect, the present application further provides a computer device. The computer device includes a memory and a processor, the memory stores a computer program, and the processor implements the steps in the image recognition method based on deep learning as described in any embodiment of the first aspect when executing the computer program.

[0040] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps in the image recognition method based on deep learning as described in any one of the embodiments in the first aspect.

[0041] In a fifth aspect, the present application further provides a computer program product. The computer program product includes a computer program, and when the computer program is executed by a processor, the steps in the image recognition method based on deep learning as described in any embodiment of the first aspect are implemented.

[0042] The above-mentioned image recognition method, device, computer equipment, storage medium and computer program product based on deep learning can achieve the beneficial effects of the technical problems in the corresponding background technology by deducing the technical features in the exclusive rights:

[0043] Before implementation, the initial feature recognition network model is obtained, and the target processing unit that meets the preset removal index is screened out from the processing units of the initial feature recognition network model; the target processing unit is removed from the initial feature recognition network model to obtain the target feature recognition network model. In the application, the image to be processed is obtained, and the image to be processed is processed based on a preset preprocessing process to obtain the target image; the target image is input into the target feature recognition network model for processing to obtain the recognition result, and the processing unit of the target feature recognition network model is less than the initial feature recognition network model; the feature information of the feature unit in the image to be processed is determined according to the recognition result. Compared with the defects in the prior art, by removing the processing units with less accuracy in the model, the calculation steps of the model for image recognition are reduced, and the image recognition accuracy is guaranteed as a whole while improving the processing efficiency of image recognition. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 A diagram of an application environment of an image recognition method based on deep learning in one embodiment;

[0045] Figure 2 A first flow chart of an image recognition method based on deep learning in one embodiment;

[0046] Figure 3 is a structural block diagram of an image recognition device based on deep learning in one embodiment;

[0047] Figure 4 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0048] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0049] The present invention provides an image recognition method based on deep learning, which can be applied to Figure 1In the application environment shown. Among them, the terminal 102 communicates with the server 104 through the network. The data storage system can store the data that the server 104 needs to process. The data storage system can be integrated on the server 104, or it can be placed on the cloud or other network servers. Among them, the terminal 102 can be but is not limited to various personal computers, laptops, smart phones, tablets, Internet of Things devices and portable wearable devices. The Internet of Things devices can be smart speakers, smart TVs, smart air conditioners, smart car-mounted devices, etc. Portable wearable devices can be smart watches, smart bracelets, head-mounted devices, etc. The server 104 can be implemented with an independent server or a server cluster consisting of multiple servers.

[0050] In one embodiment, Figure 2 As shown, a deep learning-based image recognition method is provided, which is applied to Figure 1 The terminal in is used as an example to illustrate, including the following steps:

[0051] Step 202: Acquire an image to be processed, and process the image to be processed based on a preset preprocessing process to obtain a target image;

[0052] Step 204: 1. Input the target image into a target feature recognition network model for processing to obtain a recognition result, wherein the target feature recognition network model has fewer processing units than the initial feature recognition network model;

[0053] Step 206: Determine feature information of feature units in the image to be processed according to the recognition result.

[0054] In the above-mentioned image recognition method based on deep learning, reasonable deduction is performed in combination with the technical features in the embodiment to achieve the beneficial effect of solving the technical problems raised in the background technology:

[0055] Before implementation, the initial feature recognition network model is obtained, and the target processing unit that meets the preset removal index is screened out from the processing units of the initial feature recognition network model; the target processing unit is removed from the initial feature recognition network model to obtain the target feature recognition network model. In the application, the image to be processed is obtained, and the image to be processed is processed based on a preset preprocessing process to obtain the target image; the target image is input into the target feature recognition network model for processing to obtain the recognition result, and the processing unit of the target feature recognition network model is less than the initial feature recognition network model; the feature information of the feature unit in the image to be processed is determined according to the recognition result. Compared with the defects in the prior art, by removing the processing units with less accuracy in the model, the calculation steps of the model for image recognition are reduced, and the image recognition accuracy is guaranteed as a whole while improving the processing efficiency of image recognition.

[0056] In one embodiment, the step 204 includes:

[0057] Step 302: obtaining an initial feature recognition network model, and selecting target processing units that meet a preset removal index from the processing units of the initial feature recognition network model;

[0058] Step 304: remove the target processing unit from the initial feature recognition network model to obtain the target feature recognition network model.

[0059] In one embodiment, step 302 includes:

[0060] Step 402: Acquire a historical sample set, where the historical sample set includes paired historical image samples and historical image feature information;

[0061] Step 404: traverse the processing unit of the initial feature recognition network model without the processing unit, and verify the accuracy of the result of the initial feature recognition network model after the processing unit is removed through the historical sample set;

[0062] Step 406: Determine the target processing unit according to the verification result of the result accuracy.

[0063] In one embodiment, the method further comprises:

[0064] Step 502: In response to different feature recognition requirements, obtaining the historical sample set corresponding to the feature recognition requirements;

[0065] Step 504: Reacquire the target feature recognition network model based on the updated historical sample set.

[0066] In one of the embodiments, the target processing unit includes a combination of one or more of a model layer, a model module, a model channel, and a model calculation parameter.

[0067] In one embodiment, the preprocessing process includes one or more combinations of denoising processing, image enhancement processing, resizing processing, color space conversion processing, edge detection processing, and image smoothing processing.

[0068] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0069] Based on the same inventive concept, the embodiment of the present application also provides a deep learning-based image recognition device for implementing the deep learning-based image recognition method mentioned above. The implementation solution provided by the device to solve the problem is similar to the implementation solution recorded in the above method, so the specific limitations in one or more of the embodiments of the deep learning-based image recognition device provided below can refer to the above limitations on the deep learning-based image recognition method, and will not be repeated here.

[0070] In one embodiment, Figure 3 As shown, a deep learning-based image recognition device is provided, comprising: a preprocessing module, a model recognition module and a feature information module, wherein:

[0071] A preprocessing module is used to obtain an image to be processed, and process the image to be processed based on a preset preprocessing process to obtain a target image;

[0072] A model recognition module, used for inputting the target image into a target feature recognition network model for processing to obtain a recognition result, wherein the processing unit of the target feature recognition network model is less than that of the initial feature recognition network model;

[0073] The feature information module is used to determine the feature information of the feature unit in the image to be processed according to the recognition result.

[0074] In one embodiment, the model identification module includes:

[0075] A processing unit screening module, used to obtain an initial feature recognition network model, and screen out target processing units that meet preset removal indicators from the processing units of the initial feature recognition network model;

[0076] The model pruning module is used to remove the target processing unit from the initial feature recognition network model to obtain the target feature recognition network model.

[0077] In one embodiment, the processing unit screening module includes:

[0078] A historical sample set module, used to obtain a historical sample set, wherein the historical sample set includes paired historical image samples and historical image feature information;

[0079] An accuracy verification module, used for traversing the processing unit of the initial feature recognition network model after removing the processing unit, and verifying the accuracy of the result of the initial feature recognition network model after removing the processing unit through the historical sample set;

[0080] The target processing unit module is used to determine the target processing unit according to the verification result of the result accuracy.

[0081] In one embodiment, the device further comprises:

[0082] An identification requirement module, used for obtaining the historical sample set corresponding to the feature identification requirement in response to different feature identification requirements;

[0083] A model updating module is used to reacquire the target feature recognition network model based on the updated historical sample set.

[0084] In one of the embodiments, the target processing unit includes a combination of one or more of a model layer, a model module, a model channel, and a model calculation parameter.

[0085] In one embodiment, the preprocessing process includes one or more combinations of denoising processing, image enhancement processing, resizing processing, color space conversion processing, edge detection processing, and image smoothing processing.

[0086] Each module in the above-mentioned image recognition device based on deep learning can be implemented in whole or in part by software, hardware and a combination thereof. Each of the above-mentioned modules can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute the operations corresponding to each of the above modules.

[0087] In one embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as follows: Figure 4As shown. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit and an input device. Among them, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface, the display unit and the input device are connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and the external device. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be implemented through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. When the computer program is executed by the processor, a method for image recognition based on deep learning is implemented. The display unit of the computer device is used to form a visually visible picture, which can be a display screen, a projection device or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad set on the computer device shell, or an external keyboard, touchpad or mouse.

[0088] Those skilled in the art will understand that Figure 4 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0089] In one embodiment, a computer device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above method embodiments when executing the computer program.

[0090] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0091] In one embodiment, a computer program product is provided, including a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0092] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant laws, regulations and standards of relevant countries and regions.

[0093] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.

[0094] The technical features of the above embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0095] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. An image recognition method based on deep learning, characterized in that: The method comprises: Acquire an image to be processed, and process the image to be processed based on a preset preprocessing process to obtain a target image; Inputting the target image into a target feature recognition network model for processing to obtain a recognition result, wherein the processing unit of the target feature recognition network model is less than that of the initial feature recognition network model; Determine feature information of feature units in the image to be processed according to the recognition result.

2. The method according to claim 1, characterized in that The inputting the target image into the target feature recognition network model for processing to obtain the recognition result comprises: Acquire the initial feature recognition network model, and select target processing units that meet preset removal indicators from the processing units of the initial feature recognition network model; The target processing unit is removed from the initial feature recognition network model to obtain the target feature recognition network model.

3. The method according to claim 2, characterized in that The step of obtaining an initial feature recognition network model and selecting a target processing unit that meets a preset removal index from the processing units of the initial feature recognition network model comprises: Acquire a historical sample set, wherein the historical sample set includes paired historical image samples and historical image feature information; Traversing the processing unit of the initial feature recognition network model without the processing unit, and verifying the accuracy of the result of the initial feature recognition network model after the processing unit is removed through the historical sample set; The target processing unit is determined according to the verification result of the result accuracy.

4. The method according to claim 3, characterized in that The method further comprises: In response to different feature recognition requirements, obtaining the historical sample set corresponding to the feature recognition requirements; The target feature recognition network model is reacquired based on the updated historical sample set.

5. The method according to any one of claims 1 to 4, characterized in that: The target processing unit includes one or more combinations of model layers, model modules, model channels, and model calculation parameters.

6. The method according to claim 1, characterized in that The preprocessing process includes one or more combinations of denoising processing, image enhancement processing, size adjustment processing, color space conversion processing, edge detection processing and image smoothing processing.

7. An image recognition device based on deep learning, characterized in that: The device comprises: A preprocessing module is used to obtain an image to be processed, and process the image to be processed based on a preset preprocessing process to obtain a target image; A model recognition module, used for inputting the target image into a target feature recognition network model for processing to obtain a recognition result, the target feature recognition network model; The feature information module is used to determine the feature information of the feature unit in the image to be processed according to the recognition result.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images