Image recognition method and system, electronic equipment and storage medium

Abstract

The invention discloses an image recognition method and system, electronic equipment and a storage medium, and relates to the technical field of image recognition, and the method comprises the steps: carrying out the recognition of a preset image through a target recognition model; and when the expected target is identified from the preset image, identifying the local image where the expected target is located by using the state identification model, and determining the state of the expected target. According to the method, the target recognition model and the state recognition model are more targeted, the expected target and the state of the expected target can be recognized more precisely by recognizing the expected target and the state of the expected target step by step, and a user can make a decision more precisely according to the expected target and the state of the expected target.

Description

Background Art

[0002] At present, when performing image recognition, the same model is often used to identify the target in the image and identify some information of the target, but for images of low quality, the recognition results are often not accurate enough. Summary of the invention

[0003] The technical problem to be solved by the present invention is to address the deficiencies of the prior art and specifically provide an image recognition method, system, electronic device and storage medium, as follows:

[0004] 1) In the first aspect, the present invention provides an image recognition method, and the specific technical solution is as follows:

[0005] Use the target recognition model to identify the preset image;

[0006] When a desired target is identified from a preset image, a state recognition model is used to identify the local image where the desired target is located to determine the state of the desired target.

[0007] The beneficial effects of an image recognition method provided by the present invention are as follows:

[0008] The target recognition model and the state recognition model are more targeted. By identifying the desired target and the state of the desired target in steps, the desired target and the state of the desired target can be more accurately identified. Users can make more accurate decisions based on the desired target and the state of the desired target.

[0009] Based on the above solution, an image recognition method of the present invention can also be improved as follows.

[0010] Furthermore, the target recognition model and the state recognition model are obtained by training different deep learning models.

[0011] Furthermore, the state recognition model is obtained by training the Efficientnet model, and the target recognition model is obtained by training the convolutional neural network, the fast regional convolutional neural network, the long short-term memory network or the Yolo model.

[0012] Furthermore, it also includes:

[0013] In the preset image, mark the desired target and mark the category of the desired target.

[0014] The beneficial effect of adopting the above further solution is that it is convenient for users to view the desired target and the category of the desired target in the preset image, which is more intuitive.

[0015] 2) In the second aspect, the present invention also provides an image recognition system, and the specific technical solution is as follows:

[0016] It includes a target recognition module and a state recognition module;

[0017] The target recognition module is used to: recognize the preset image using the target recognition model;

[0018] The state recognition module is used to: when a desired target is recognized from a preset image, use the state recognition model to recognize the local image where the desired target is located to determine the state of the desired target.

[0019] Based on the above solution, an image recognition system of the present invention can also be improved as follows.

[0020] Furthermore, the target recognition model and the state recognition model are obtained by training different deep learning models.

[0021] Furthermore, the state recognition model is obtained by training the Efficientnet model, and the target recognition model is obtained by training the convolutional neural network, the fast regional convolutional neural network, the long short-term memory network or the Yolo model.

[0022] Furthermore, it also includes a marking module, which is used to: mark the desired target in the preset image and mark the category of the desired target.

[0023] 3) In a third aspect, the present invention further provides an electronic device, comprising a processor, the processor being coupled to a memory, the memory storing at least one computer program, the at least one computer program being loaded and executed by the processor, so that the electronic device implements any one of the above-mentioned image recognition methods.

[0024] 4) In a fourth aspect, the present invention further provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, any of the above-mentioned image recognition methods is implemented.

[0025] It should be noted that the beneficial effects achieved by the technical solutions of the second to fourth aspects of the present invention and the corresponding possible implementation methods can be found in the above-mentioned technical effects of the first aspect and its corresponding possible implementation methods, and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments of the present invention:

[0027] Figure 1 A schematic diagram of a flow chart of an image recognition method according to an embodiment of the present invention;

[0028] Figure 2 A schematic diagram of the structure of an image recognition system according to an embodiment of the present invention;

[0029] Figure 3 The figure is a schematic diagram of the structure of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0030] The principles and features of the present invention are described below. The examples given are only used to explain the present invention and are not used to limit the scope of the present invention.

[0031] The technical solution of the present invention and how the technical solution of the present invention solves the above-mentioned technical problems are described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present invention will be described below in conjunction with the accompanying drawings.

[0032] like Figure 1 As shown, an image recognition method according to an embodiment of the present invention comprises the following steps:

[0033] S1. Using the target recognition model to recognize the preset image;

[0034] S2. When a desired target is identified from a preset image, a state recognition model is used to identify the local image where the desired target is located to determine the state of the desired target.

[0035] Optionally, in the above technical solution, the target recognition model and the state recognition model are obtained by training different deep learning models.

[0036] Optionally, in the above technical solution, the state recognition model is obtained by training the Efficientnet model, and the target recognition model is obtained by training the convolutional neural network, the fast regional convolutional neural network, the long short-term memory network or the Yolo model.

[0037] Optionally, in the above technical solution, it also includes:

[0038] S3. Mark the desired target in the preset image and mark the category of the desired target. This makes it easier for the user to view the desired target and the category of the desired target in the preset image, which is more intuitive.

[0039] An image recognition method of the present invention is explained through the following embodiments.

[0040] Embodiment 1:

[0041] In Example 1, the preset image is a medical image, the desired target is: lesion, and the category of the desired target is: severity of lesion, wherein the lesion may specifically be a lesion of spinal disease, or a lesion of other diseases, and the severity of the lesion may be divided according to actual conditions.

[0042] Embodiment 1 specifically comprises the following steps:

[0043] S101. Collect multiple medical images of spinal diseases, and preprocess each medical image, wherein the preprocessing includes at least one of grayscale conversion, noise removal, image enhancement, geometric transformation, and normalization. Specifically:

[0044] 1) Grayscale conversion: Medical images are converted into grayscale images using grayscale conversion methods including weighted averaging to reduce the amount of calculation while retaining sufficient feature information.

[0045] 2) Noise removal: Use filters (such as mean filters, median filters, or Gaussian filters) to remove noise from medical images.

[0046] 3) Image enhancement: Use Laplacian operator, gradient operator, etc. to enhance the edges and details of medical images, and improve the contrast of medical images through histogram equalization, adaptive histogram equalization and other methods.

[0047] 4) Geometric transformation: Correct the geometric distortion in medical images through scaling, rotation, translation, affine transformation, etc., and use perspective transformation to deal with the deformation of medical images caused by perspective projection.

[0048] 5) Normalization: Through Z-score normalization or Min-Max normalization, the pixel values ​​of medical images are adjusted to a specific range (such as 0-1 or -1 to 1), which helps in subsequent model training.

[0049] 6) Data enhancement: Perform random transformations on medical images (such as rotation, flipping, cropping, color jittering, etc.) to increase the diversity of medical images and improve the generalization ability of the trained state recognition model.

[0050] S102, segmenting an image including only the lesion area from each preprocessed medical image and marking the severity level, each image including only the lesion area and the corresponding severity level is a first training sample, and multiple first training samples constitute a first training set.

[0051] S103: Train the EfficientNet model based on the first training set to obtain a trained EfficientNet model, namely, a state recognition model.

[0052] S104. Train the Yolo model based on the second training set to obtain a trained Yolo model, i.e., a target recognition model, wherein the location of the lesion in each preprocessed medical image is marked, and each preprocessed medical image and the corresponding marked lesion location constitute a second training sample, and multiple second training samples constitute a second training set.

[0053] S105. Use the target recognition model (the trained Yolo model in S104) to recognize the medical image (preset image) specified by the user. When the lesion of the spinal disease is identified from the medical image specified by the user, perform image segmentation on the medical image specified by the user to obtain a local image where the lesion of the spinal disease is located.

[0054] S106. Using the state recognition model (the trained EfficientNet model in S103), the local image where the spinal disease lesions are located is identified to determine the severity of the spinal disease lesions.

[0055] S107. Mark the lesions of the spinal disease in the medical image specified by the user, and mark the severity of the lesions of the spinal disease. This is convenient for users (such as doctors, etc.) to view the lesions of the spinal disease and the severity of the lesions of the spinal disease in the specified medical image, which is more intuitive and accurate, and helps users (such as doctors, etc.) to carry out subsequent processing (such as diagnosis and correction of treatment plans, etc.), and can effectively solve the problem that "the existing technology fails to effectively combine the location of the lesion with the severity, resulting in inaccurate diagnosis results".

[0056] Embodiment 2:

[0057] In Embodiment 2, the preset image is: an image of a target road captured by a camera, the desired target is: a vehicle on the target road, and the categories of the desired target are: violation and non-violation.

[0058] Embodiment 2 specifically comprises the following steps:

[0059] S201. Collect multiple road images and preprocess each road image, where the preprocessing includes at least one of grayscale conversion, noise removal, image enhancement, geometric transformation, and normalization.

[0060] S202. Segment an image including only vehicles from each preprocessed road image, and annotate a label of each vehicle. The labels are divided into: vehicle violation and vehicle non-violation. When the vehicle violates the law, the label also includes the reason for the vehicle violation. Each image including only one vehicle and the corresponding label constitute a third training sample, and multiple third training samples constitute a third training set.

[0061] S203, training the Efficientnet model based on the third training set to obtain a trained Efficientnet model, i.e., a state recognition model;

[0062] S204. Train the Yolo model based on the fourth training set to obtain a trained Yolo model, i.e., a target recognition model, wherein vehicles in each preprocessed road image are labeled, and each preprocessed road image and the corresponding labeled vehicle respectively constitute a fourth training sample, and multiple fourth training samples constitute a fourth training set.

[0063] S205. Use the target recognition model (the trained Yolo model in S204) to recognize the road image (preset image) specified by the user. When a vehicle is recognized from the road image specified by the user, perform image segmentation on the road image specified by the user to obtain a local image where each vehicle is located.

[0064] S206. Use the state recognition model (the trained Efficientnet model in S203) to identify the local image where each vehicle is located to determine whether each vehicle violates the rules. When any vehicle violates the rules, the reason for the violation will be output.

[0065] It also includes: when a violating vehicle appears, real-time tracking of the violating vehicle is carried out, specifically:

[0066] S207: Acquire multiple real-time road images, and use the trained license plate area model to identify the license plate area image of each vehicle in each real-time road image.

[0067] It should be noted that the trained license plate area model only recognizes the license plate area, and does not need to recognize the specific license plate number. It can improve the recognition accuracy of the license plate area, and can greatly reduce the probability of tracking failure due to damage or occlusion of the license plate number in subsequent matching. It can also greatly reduce the amount of data calculation and improve data processing efficiency. In addition, during the training process, fewer labels will be annotated, which can reduce manpower and material costs. The trained license plate area model can be a trained convolutional neural network, etc., which can be set according to actual conditions.

[0068] S208, by using an interpolation algorithm or the like, the license plate area image of each vehicle in each real-time road image and the license plate area image of the illegal vehicle in S206 are converted into images of the same size, and then are converted into binary images.

[0069] S209. In each binary image, with the central pixel of the binary image as the center, every W pixels are expanded outward layer by layer in a rectangular or circular manner to obtain multiple layers of pixels. Each layer of pixels forms a rectangle or a circle. Black pixels in the binary image are recorded as 0, and white pixels are recorded as 1. Multiple layers of pixels form a matrix. The similarity between the matrix corresponding to each vehicle and the matrix corresponding to the illegal vehicle in each real-time road image is calculated, and the vehicle corresponding to the maximum similarity in each real-time road image is determined as the illegal vehicle. In this way, real-time tracking of the illegal vehicle is achieved. Moreover, compared with the overall image similarity comparison in the prior art, the amount of calculation can be reduced while ensuring the tracking accuracy by binarizing the image and extracting some pixels. W can be set according to the actual situation.

[0070] In another embodiment, multiple layers of pixel points in each binary image are arranged according to the same rule to obtain a sequence corresponding to each binary image, the sequence consisting of 1 and 0, and each sequence is regarded as one-dimensional time series data. Then, the similarity between the one-dimensional time series data corresponding to each vehicle in each real-time road image and the one-dimensional time series data corresponding to the illegal vehicle is calculated, and the vehicle corresponding to the maximum similarity in each real-time road image is determined as the illegal vehicle, thereby realizing real-time tracking of the illegal vehicles.

[0071] The process of calculating the similarity between the one-dimensional time series data corresponding to each vehicle in each real-time road image and the one-dimensional time series data corresponding to the illegal vehicle includes:

[0072] Calculate the distances between all points in the two sequences and construct a distance matrix. Use dynamic programming to find the best alignment between the two sequences. Based on the best alignment, calculate a cumulative distance matrix. Starting from the lower right corner of the cumulative distance matrix, backtrack and select the path with the minimum cumulative distance. Based on the last element of the cumulative distance matrix, get the similarity between the two sequences.

[0073] Among them, the dynamic programming method is used to find the best alignment between two sequences, including:

[0074] A two-dimensional array OPT is used, where OPT[i][j] represents the optimal alignment cost between the first i characters of sequence X and the first j characters of sequence Y, where sequence X represents the sequence corresponding to any vehicle in any real-time road image, and the following preset conditions are defined:

[0075] 1) If X[i-1] and Y[j-1] are equal, the alignment cost will increase by a small positive number (such as 0.01 or 0.02, indicating the matching score);

[0076] 2) If a character needs to be inserted into X to align with Y, the alignment cost is increased by a large value (such as 10 or 20, etc., indicating the penalty for the insertion operation);

[0077] 3) If a character in X has no corresponding character in Y, the character needs to be deleted, and the alignment cost will also increase by a larger value (such as 10 or 20, etc., indicating the penalty for the deletion operation).

[0078] Starting from OPT[1][1], the remaining elements of the OPT array are iteratively calculated in combination with the above preset conditions until OPT[n][m], where n and m are the lengths of X and Y respectively, and the best alignment of the two sequences is constructed based on the backtracking path. This usually involves inserting spaces or deleting characters when necessary to align the two sequences.

[0079] In the above embodiments, although the steps are numbered S1, S2, etc., these are only specific embodiments given by the present invention. Those skilled in the art may adjust the execution order of S1, S2, etc. according to actual conditions, which is also within the protection scope of the present invention. It can be understood that in some embodiments, some or all of the above embodiments may be included.

[0080] like Figure 2 As shown, an image recognition system 200 according to an embodiment of the present invention includes a target recognition module 201 and a state recognition module 202;

[0081] The target recognition module 201 is used to: recognize a preset image using a target recognition model;

[0082] The state recognition module 202 is used to: when a desired target is recognized from a preset image, recognize the local image where the desired target is located using a state recognition model to determine the state of the desired target.

[0083] Optionally, in the above technical solution, the target recognition model and the state recognition model are obtained by training different deep learning models.

[0084] Optionally, in the above technical solution, the state recognition model is obtained by training the Efficientnet model, and the target recognition model is obtained by training the convolutional neural network, the fast regional convolutional neural network, the long short-term memory network or the Yolo model.

[0085] Optionally, the above technical solution further includes a marking module, which is used to: mark the desired target in a preset image and mark the category of the desired target.

[0086] It should be noted that the beneficial effects of the image recognition system 200 provided in the above embodiment are the same as the beneficial effects of the above image recognition method, which will not be described in detail here. In addition, when the system provided in the above embodiment realizes its functions, only the division of the above functional modules is used as an example. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the system can be divided into different functional modules according to actual conditions to complete all or part of the functions described above. In addition, the system and method embodiments provided in the above embodiment belong to the same concept, and the specific implementation process is detailed in the method embodiment, which will not be described in detail here.

[0087] Among them, the image recognition system of the present invention can be a computer program (including program code) running in a computer device. For example, the image recognition system of the present invention is an application software that can be used to execute the corresponding steps in the image recognition method of the present invention.

[0088] In some embodiments, the image recognition system of the present invention can be implemented by a combination of software and hardware. As an example, the image recognition system of the present invention can be a processor in the form of a hardware decoding processor, which is programmed to execute the image recognition method of the present invention. For example, the processor in the form of a hardware decoding processor can adopt one or more application-specific integrated circuits (ASICs), DSPs, programmable logic devices (PLDs), complex programmable logic devices (CPLDs), field-programmable gate arrays (FPGAs) or other electronic components.

[0089] The modules involved in the embodiments of the present invention may be implemented in software or hardware. The name of a module does not limit the module itself in some cases.

[0090] An electronic device according to an embodiment of the present invention includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, any one of the above-mentioned image recognition methods is implemented. That is to say, an electronic device according to an embodiment of the present invention may include but is not limited to: a processor and a memory; a memory for storing a computer program; a processor for executing an image recognition method shown in any embodiment of the present invention by calling a computer program.

[0091] In an alternative embodiment, an electronic device is provided, such as Figure 3 As shown, Figure 3 The electronic device 4000 shown includes: a processor 4001 and a memory 4003. The processor 4001 and the memory 4003 are connected, such as through a bus 4002. Optionally, the electronic device 4000 may also include a transceiver 4004, which may be used for data interaction between the electronic device and other electronic devices, such as data transmission and / or data reception. It should be noted that in actual applications, the transceiver 4004 is not limited to one, and the structure of the electronic device 4000 does not constitute a limitation on the embodiments of the present invention.

[0092] Processor 4001 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array) or other programmable logic devices, transistor logic devices, hardware components or any combination thereof. It may implement or execute various exemplary logic blocks, modules and circuits described in conjunction with the disclosure of the present invention. Processor 4001 may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc.

[0093] The bus 4002 may include a path to transmit information between the above components. The bus 4002 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus. The bus 4002 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 3 In the figure, only one thick line is used to represent the bus 4002, but this does not mean that there is only one bus or one type of bus.

[0094] The memory 4003 can be a ROM (Read Only Memory) or other types of static storage devices that can store static information and instructions, a RAM (Random Access Memory) or other types of dynamic storage devices that can store information and instructions, or an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read Only Memory) or other optical disk storage, optical disk storage (including compressed optical disk, laser disk, optical disk, digital versatile disk, Blu-ray disk, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited to this.

[0095] The memory 4003 is used to store the application code (computer program) for executing the solution of the present invention, and the execution is controlled by the processor 4001. The processor 4001 is used to execute the application code stored in the memory 4003 to implement the content shown in the above method embodiment.

[0096] Among them, the electronic device can also be a terminal device, and the terminal device can be any device that can install applications, including at least one of a smart phone, a tablet computer, a laptop computer, a desktop computer, a smart speaker, a smart watch, a smart TV, and a smart car device.

[0097] It should be noted that Figure 3 The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0098] A computer-readable storage medium according to an embodiment of the present invention stores a computer program, and when the computer program is executed by a processor, any of the above-mentioned image recognition methods is implemented.

[0099] Optionally, the computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a compact disc (CD-ROM), a magnetic tape, a floppy disk, an optical data storage device, and the like.

[0100] In an exemplary embodiment, a computer program product or a computer program is also provided, the computer program product or the computer program comprising computer instructions, the computer instructions being stored in a computer-readable storage medium. A processor of an electronic device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the electronic device performs any of the above-mentioned image recognition methods.

[0101] Computer program code for performing the operations of the present invention may be written in one or more programming languages, or a combination thereof, including object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., through the Internet using an Internet service provider).

[0102] It should be understood that the flow charts and block diagrams in the accompanying drawings illustrate the possible implementation architecture, functions and operations of the method and computer program product according to various embodiments of the present invention. In this regard, each box in the flow chart or block diagram can represent a module, a program segment, or a part of a code, and the module, the program segment, or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a sequence different from that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flow chart, and the combination of the boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0103] The computer-readable storage medium provided by the embodiment of the present invention may be, but is not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or component, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in conjunction with an instruction execution system, device, or component.

[0104] The computer-readable storage medium carries one or more programs. When the one or more programs are executed by the electronic device, the electronic device executes the method shown in the above embodiment.

[0105] The above description is only a preferred embodiment of the present invention and an explanation of the technical principles used. Those skilled in the art should understand that the disclosure scope involved in the present invention is not limited to the technical solution formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above disclosed concept. For example, the above features are replaced with the technical features with similar functions disclosed in the present invention (but not limited to) by each other.

[0106] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application are used to distinguish similar objects and represent the definition of a specific order or sequence. The order of use of similar objects can be interchanged where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than the order shown or described.

[0107] Those skilled in the art know that the present invention can be implemented as a system, method or computer program product. Therefore, the present invention can be specifically implemented in the following forms, namely: it can be complete hardware, it can be complete software (including firmware, resident software, microcode, etc.), or it can be a combination of hardware and software, which is generally referred to as "circuit", "module" or "system" herein. In addition, in some embodiments, the present invention can also be implemented in the form of a computer program product in one or more computer-readable media, and the computer-readable medium contains computer-readable program code.

[0108] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present invention. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present invention.

Claims

1. An image recognition method, characterized in that: include: Use the target recognition model to identify the preset image; When a desired target is identified from the preset image, the state recognition model is used to identify the local image where the desired target is located to determine the state of the desired target.

2. An image recognition method according to claim 1, characterized in that: The target recognition model and the state recognition model are obtained by training different deep learning models.

3. An image recognition method according to claim 2, characterized in that: The state recognition model is obtained by training the Efficientnet model, and the target recognition model is obtained by training the convolutional neural network, the fast regional convolutional neural network, the long short-term memory network or the Yolo model.

4. An image recognition method according to any one of claims 1 to 3, characterized in that: Also includes: In the preset image, the desired target is marked, and the category of the desired target is marked.

5. An image recognition system, characterized in that: It includes a target recognition module and a state recognition module; The target recognition module is used to: recognize a preset image using a target recognition model; The state recognition module is used to: when a desired target is recognized from the preset image, use a state recognition model to recognize the local image where the desired target is located, and determine the state of the desired target.

6. An image recognition system according to claim 5, characterized in that: The target recognition model and the state recognition model are obtained by training different deep learning models.

7. An image recognition system according to claim 6, characterized in that: The state recognition model is obtained by training the Efficientnet model, and the target recognition model is obtained by training the convolutional neural network, the fast regional convolutional neural network, the long short-term memory network or the Yolo model.

8. An image recognition system according to any one of claims 5 to 7, characterized in that: It also includes a marking module, which is used to mark the desired target in the preset image and mark the category of the desired target.

9. An electronic device, characterized in that: The invention comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements an image recognition method as claimed in any one of claims 1 to 4 when executing the computer program.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the image recognition method according to any one of claims 1 to 4 is implemented.

Images