Target object identification method and device, equipment, medium and product

Abstract

The invention discloses a target object identification method and device, equipment, a medium and a product, and the method comprises the steps: obtaining a to-be-processed image containing a target object; performing multi-layer convolution processing on the to-be-processed image to obtain a to-be-recognized image group corresponding to the to-be-processed image; and performing feature coding processing on all the images in the to-be-recognized image group to obtain a target histogram, and performing object recognition on the target object based on the target histogram. The problem that the target object cannot be accurately recognized if the shielding information exists on the face of the target object in the face recognition process of the target object in the prior art is solved, and the effect of accurately recognizing the target object even if the shielding information exists on the face of the target object is achieved.

Description

Technical Field

[0001] The present invention relates to the field of image recognition technology, and in particular to a method, device, equipment, medium and product for identifying a target object. Background Art

[0002] When providing business services to users, it is necessary to identify the user's identity in order to ensure the security of the user's information.

[0003] Currently, users can be identified by performing facial recognition on images containing facial information. In existing technologies, the difficulty of user identification is increased due to factors such as light intensity, facial posture, or occlusion of the user's face. This is because when facial recognition is performed based on traditional methods, the facial geometric feature vector structure is relatively simple, and the available feature information is limited.

[0004] In order to accurately recognize the user's face, the face recognition method needs to be improved. Summary of the Invention

[0005] The present invention provides a target object recognition method, device, equipment, medium and product to solve the problem in the prior art of being unable to accurately recognize the target object when there is occlusion information on the target object's face during facial recognition.

[0006] In a first aspect, an embodiment of the present invention provides a method for identifying a target object, comprising:

[0007] Obtain an image to be processed containing a target object;

[0008] Performing multi-layer convolution processing on the image to be processed to obtain a group of images to be identified corresponding to the image to be processed;

[0009] Perform feature coding processing on all images in the to-be-identified image group to obtain a target histogram, and perform object recognition on the target object based on the target histogram.

[0010] In a second aspect, an embodiment of the present invention further provides a target object recognition device, comprising:

[0011] An acquisition module, used for acquiring an image to be processed containing a target object;

[0012] An image group determination module is used to perform multi-layer convolution processing on the image to be processed to obtain an image group to be identified corresponding to the image to be processed;

[0013] The object recognition module is used to perform feature coding processing on all images in the image group to be recognized to obtain a target histogram, and perform object recognition on the target object based on the target histogram.

[0014] In a third aspect, an embodiment of the present invention further provides an electronic device, including:

[0015] at least one processor; and

[0016] a memory communicatively connected to the at least one processor; wherein,

[0017] The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can perform the target object recognition method described in any embodiment of the present invention.

[0018] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the target object recognition method described in any embodiment of the present invention when executed.

[0019] In a fifth aspect, an embodiment of the present invention further provides a computer program product, comprising a computer program, which, when executed by a processor, implements a target object recognition method as described in any one of the embodiments of the present invention.

[0020] The technical solution of the embodiment of the present invention is to obtain an image to be processed containing a target object; perform multi-layer convolution processing on the image to be processed to obtain a group of images to be identified corresponding to the image to be processed; perform feature encoding processing on all images in the group of images to be identified to obtain a target histogram, and perform object recognition on the target object based on the target histogram. In this technical solution, the group of images to be identified corresponding to the image to be processed is obtained by performing convolution processing on the image to be processed twice, wherein the convolution kernel in the first convolution is a convolution kernel constructed based on the image to be processed, and the convolution kernel in the second convolution is a convolution kernel constructed based on a random image among the multiple images generated by the first convolution. Furthermore, a binary hash coding process is performed on the group of images to be identified to obtain a target histogram, so as to highlight the structure and features of the target object based on the target histogram, and then the target object is recognized by analyzing and processing the feature information in the target histogram based on the classifier. The present invention solves the problem in the prior art that, during the face recognition process of the target object, the target object cannot be accurately identified if there is occlusion information on the face of the target object, and achieves the effect of accurately identifying the target object even when there is occlusion information on the face of the target object.

[0021] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present invention, nor is it intended to limit the scope of the present invention. Other features of the present invention will become readily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0023] Figure 1 This is a flowchart of a target object identification method provided in accordance with the first embodiment of the present invention;

[0024] Figure 2 This is a flow chart of a target object identification method provided in accordance with the second embodiment of the present invention;

[0025] Figure 3 2 is a schematic diagram of performing multi-layer convolution on an image to be processed according to a second embodiment of the present invention;

[0026] Figure 4 This is a schematic structural diagram of a target object recognition device provided in accordance with a third embodiment of the present invention;

[0027] Figure 5 It is a structural diagram of an electronic device for implementing the target object recognition method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0028] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only embodiments of a part of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention. The acquisition, transmission, storage, use, and processing of data in the technical solution of this application comply with the relevant provisions of national laws and regulations. It should be noted that in the embodiments of this application, certain software, components, or models that already exist in the industry may be mentioned. They should be regarded as exemplary. Their purpose is only to illustrate the feasibility of the implementation of the technical solution of this application, but it does not mean that the applicant has or must use the solution.

[0029] It should be noted that the terms "first," "second," and the like in the description and claims of the present invention and the accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or precedence. It should be understood that the terms used in this manner are interchangeable where appropriate, such that the embodiments of the present invention described herein can be practiced in an order other than that illustrated or described herein.

[0030] Before elaborating on this technical solution in detail, a brief introduction to the application scenarios of this technical solution is first given to facilitate a clearer understanding of this technical solution.

[0031] The financial industry, with its highly standardized and data-intensive nature, has become a frontier for the application of artificial intelligence technology. Facial recognition technology has significant advantages in scenarios such as account opening (weak real-name electronic accounts or remote identity verification) and customer identity verification (branch reception, self-service terminals, and counter authentication), enabling it to better provide personalized services to customers. However, facial recognition technology still faces some challenges in its actual implementation. For example, light intensity and facial posture can seriously interfere with the accuracy of facial recognition; facial occlusion caused by hats, masks, or sunglasses can also affect recognition. This is mainly because traditional methods have relatively simple vector structural components for facial geometric features and limited feature expression capabilities.

[0032] Based on this, it is necessary to improve the method of target object recognition based on images.

[0033] Example 1

[0034] Figure 1 A flowchart of a target object identification method is provided for the first embodiment of the present invention. This embodiment can be applied to obtaining an image to be processed containing facial features of a target object, performing multi-layer convolution processing on the image to be processed to obtain an image group to be identified, and then performing binary hash coding processing on the image to be identified in the image group to obtain a target histogram, performing feature recognition on the target histogram based on a target classifier, and then identifying the target object. The method can be executed by a target object identification device, which can be implemented in the form of hardware and / or software, and can be configured in a computing device that can execute the target object identification method.

[0035] like Figure 1 As shown, the method includes:

[0036] S110: Acquire an image to be processed containing a target object.

[0037] The target object refers to the user whose face needs to be recognized. This can be a user with or without occlusion information. The image to be processed refers to an image of the target object's face captured by an image capture device. The image to be processed can be captured under any lighting intensity, or even with facial pose interference.

[0038] For example, the image capture device may be a built-in image capture module in a mobile terminal device, or an image acquisition module at an access control station, or an image acquisition device deployed at a business site, etc.

[0039] Specifically, taking the scenario where the target object is a business user and a business branch is identifying the business user, the business branch's image capture equipment captures the business user to obtain an image to be processed. Of course, this technical solution can be applied to various scenarios requiring user identity authentication based on facial features. The application scenarios in this technical solution are only illustrative and not limiting.

[0040] S120 , performing multi-layer convolution processing on the image to be processed to obtain a group of images to be identified corresponding to the image to be processed.

[0041] The group of images to be identified includes at least one image to be identified, and the image to be identified refers to an image generated after performing multiple convolutions on the image to be processed.

[0042] It's understandable that image convolution in deep learning produces multiple output images. For example, if we perform two layers of convolution on an image to be processed, we can obtain L1 images by convolving it with kernel P1. Furthermore, we perform a second convolution on each of these L1 images with kernel P2, yielding L2 images for each image. In other words, after two convolutions on the image to be processed, we can obtain L1*L2 images. These L1*L2 images can then be used as the set of images to be identified corresponding to the image to be processed.

[0043] S130 , performing feature coding processing on all images in the image group to be identified to obtain a target histogram, and performing object identification on the target object based on the target histogram.

[0044] In this technical solution, each image to be identified in the image group to be identified is subjected to binary hash coding processing to obtain a to-be-used histogram corresponding to each image to be identified, so as to obtain a target histogram based on the average value of all to-be-used histograms.

[0045] The so-called binary hash coding refers to a process of converting all images in the image group to be identified into hash values ​​in binary form (containing only 0 and 1).

[0046] Optionally, feature encoding processing is performed on all images in the image group to be identified to obtain a target histogram, including: performing binary hash encoding processing on each image to be identified in the image group to be identified to obtain a histogram to be used corresponding to each image to be identified; and performing histogram combination on all histograms to be used to obtain a target histogram.

[0047] Based on the above example, after two convolutions, the final output is L1*L2 images to be identified, which are then used as the image group to be identified. Furthermore, the histogram to be used corresponding to each image to be identified is determined, and the feature information in the histograms to be used corresponding to all images to be identified is superimposed. The superimposed results are averaged to obtain a target histogram. This target histogram is then further analyzed based on the target classifier to classify the target object and authenticate the target object.

[0048] The advantage of this setting is that mapping the image to be identified into a binary number of fixed length can not only remove duplicate data items and reduce the size of the data set; on the other hand, the binary expression can highlight the main structure and features of the image, which is convenient for image analysis and classification.

[0049] The technical solution of the embodiment of the present invention is to obtain an image to be processed containing a target object; perform multi-layer convolution processing on the image to be processed to obtain a group of images to be identified corresponding to the image to be processed; perform feature encoding processing on all images in the group of images to be identified to obtain a target histogram, and perform object recognition on the target object based on the target histogram. In this technical solution, the group of images to be identified corresponding to the image to be processed is obtained by performing convolution processing on the image to be processed twice, wherein the convolution kernel in the first convolution is a convolution kernel constructed based on the image to be processed, and the convolution kernel in the second convolution is a convolution kernel constructed based on a random image among the multiple images generated by the first convolution. Furthermore, a binary hash coding process is performed on the group of images to be identified to obtain a target histogram, so as to highlight the structure and features of the target object based on the target histogram, and then the target object is recognized by analyzing and processing the feature information in the target histogram based on the classifier. The present invention solves the problem in the prior art that, during the face recognition process of the target object, the target object cannot be accurately identified if there is occlusion information on the face of the target object, and achieves the effect of accurately identifying the target object even when there is occlusion information on the face of the target object.

[0050] Example 2

[0051] Figure 3 This is a flowchart of a target object recognition method provided in the second embodiment of the present invention. Optionally, multi-layer convolution processing is performed on the image to be processed to obtain a group of images to be recognized corresponding to the image to be processed and refine them.

[0052] like Figure 3 As shown, the method includes:

[0053] S210: Acquire an image to be processed containing a target object, and construct a first feature extraction operator based on the image to be processed.

[0054] The first feature extraction operator refers to a feature matrix generated after processing the image to be processed based on a feature extraction algorithm.

[0055] Exemplarily, the image to be processed is processed based on a locality preserving projection algorithm to obtain a first feature extraction operator. The so-called locality preserving projection algorithm (LPP) is a simple and efficient feature extraction algorithm that can preserve the local characteristic relationships of the matrix during the process of matrix dimensionality reduction. The goal of the locality preserving projection algorithm is to find local structural information in high-dimensional data, and the dataset obtained after the dimensionality reduction mapping of the original dataset can still maintain this structural relationship.

[0056] Optionally, constructing a first feature extraction operator based on the image to be processed includes: constructing a first adjacency graph based on the image to be processed; performing feature mapping according to each pixel node in the first adjacency graph and the adjacency edge weight corresponding to at least one adjacency edge to obtain the first feature extraction operator.

[0057] The first adjacency graph includes at least one pixel node, and each pixel node has an adjacent edge with pixel nodes within a preset neighboring range.

[0058] Specifically, the specific process of processing the image to be processed based on the locality preserving projection algorithm (LPP algorithm) to obtain the first feature extraction operator is as follows:

[0059] Input image to be processed: original space R l The node data x1,x2,…,x n ;

[0060] Output the first feature extraction operator: the low-dimensional data set corresponding to the original data after mapping.

[0061] 1. Construct the first adjacency graph: When the pixel node x i With pixel node x j When they are very close, there are two ways to connect two pixel nodes:

[0062] (a) In R l Under the Euclidean norm of the space, when ||x i -x j When ||<ε, record the pixel node x i With pixel node x j connected; otherwise not connected.

[0063] (b) If pixel node i is within the k-nearest neighbor range of pixel node j, and pixel node j is also within the k-nearest neighbor range of pixel node i, then an adjacent edge is established for the two pixel nodes, otherwise they are not connected.

[0064] 2. Generate adjacent edge weights: Assign weight values ​​to the edges between pixel nodes. The adjacent edge weights can be calculated in the following two ways:

[0065] (a) If pixel node i has an adjacent edge between pixel node j, then the adjacent edge weight W ij is 1, otherwise the adjacent edge weight W ij is 0.

[0066] (b) If pixel node i has an adjacent edge between pixel node j, the adjacent edge weight is Otherwise W ij is 0.

[0067] 3. Construct the first feature extraction operator: calculate XWX T A=λXDX T The eigenvalues ​​and eigenvectors of A are obtained, and a first feature extraction operator is obtained based on the eigenvalues ​​and eigenvectors.

[0068] S220 : Perform convolution processing on the image to be processed based on the first feature extraction operator to obtain a group of images to be used.

[0069] The image group to be used includes at least one image to be used.

[0070] In practical applications, such as Figure 3 As shown, by using the first feature extraction operator to perform convolution processing on the image to be processed, Li feature matrices (i.e., images to be used) can be obtained to obtain a group of images to be used. Among them, the number of images to be used output after the first layer of convolution can be set by oneself.

[0071] Specifically, obtain the image Xi to be processed, extract features from Xi in k1*k2 blocks, and recombine the extracted feature vectors Then find the block combinations after removing the mean Finally, the LPP algorithm is used to Projection mapping is performed to obtain the first feature extraction operator of the first convolutional layer. After the image to be processed is convolved with the first feature extraction operator, a group of images to be used is obtained, wherein the group of images to be used includes M1 images.

[0072] S230 : Construct a second feature extraction operator based on the image group to be used, and perform convolution processing on each image to be used based on the second feature extraction operator to obtain an image group to be identified.

[0073] The second feature extraction operator is a feature matrix generated after extracting features from a randomly selected image to be used based on a feature extraction algorithm.

[0074] Optionally, a second feature extraction operator is constructed based on the group of images to be used, including: selecting a target use image from at least one image to be used, and constructing a second adjacency graph based on the target use image; performing feature mapping according to each pixel node in the second adjacency graph and the adjacency edge weight corresponding to at least one adjacency edge to obtain a second feature extraction operator.

[0075] The target image is an image randomly selected from the set of images to be used. The second adjacency graph includes at least one pixel node, and each pixel node has an adjacent edge with a pixel node within a preset neighboring range.

[0076] Similar to the construction method of the first feature extraction operator, a target use image is randomly selected from at least one image to be used, and the target use image is processed based on the LPP algorithm to obtain a second feature extraction operator.

[0077] On this basis, see Figure 3 , based on the second feature extraction operator, a second convolution process is performed on each image to be used. Each image to be used corresponds to M2 images to be identified, and the image group to be identified includes M1*M2 images to be identified.

[0078] S240 , performing feature coding processing on all images in the image group to be identified to obtain a target histogram, and performing object identification on the target object based on the target histogram.

[0079] Optionally, performing object recognition on the target object based on the target histogram includes: calling a pre-built target classifier; inputting the target histogram into the target classifier to obtain a classification result corresponding to the target object, and identifying the target object based on the classification result.

[0080] The target classifier is a classifier of the optimal path based on density measurement.

[0081] Based on the above example, continue to see Figure 3 , by performing two convolution operations on the processed image to obtain the image group to be identified, and then performing binary hash coding on the processed image group to obtain the target histogram. Furthermore, the target classifier is used to identify the feature information in the target histogram and output the recognition result corresponding to the target object.

[0082] For example, facial feature information of multiple users is pre-stored in a user database. The target classifier analyzes the feature information in the target histogram to determine the facial feature information of the target object. Furthermore, the target object is compared with the facial feature information of each stored user to identify a user from the stored users that matches the target object. The user database includes the facial feature information corresponding to the target object.

[0083] The technical solution of an embodiment of the present invention obtains an image to be processed containing a target object; performs multi-layer convolution on the image to be processed to obtain a group of images to be identified corresponding to the image to be processed; performs feature encoding on all images in the group of images to be identified to obtain a target histogram; and then identifies the target object based on the target histogram. In this technical solution, the group of images to be identified is obtained by performing two convolutions on the image to be processed, wherein the convolution kernel in the first convolution is constructed based on the image to be processed, and the convolution kernel in the second convolution is constructed based on a random image from multiple images generated by the first convolution. Furthermore, a binary hash coding process is performed on the group of images to be identified to obtain a target histogram, which highlights the structure and features of the target object based on the target histogram. The target histogram is then classified using a classifier to identify the target object. This solves the problem of inaccurate identification of the target object when the target object's face is obscured during facial recognition in the prior art. This solution achieves accurate identification of the target object even when the target object's face is obscured.

[0084] Example 3

[0085] Figure 4 This is a schematic diagram of the structure of a target object recognition device provided by the third embodiment of the present invention. Figure 4 As shown, the apparatus includes: an acquisition module 310 , an image group determination module 320 and an object recognition module 330 .

[0086] The acquisition module 310 is used to acquire an image to be processed containing a target object;

[0087] An image group determination module 320 is configured to perform multi-layer convolution processing on the image to be processed to obtain an image group to be identified corresponding to the image to be processed;

[0088] The object recognition module 330 is configured to perform feature coding processing on all images in the image group to be recognized, obtain a target histogram, and perform object recognition on the target object based on the target histogram.

[0089] The technical solution of an embodiment of the present invention obtains an image to be processed containing a target object; performs multi-layer convolution on the image to be processed to obtain a group of images to be identified corresponding to the image to be processed; performs feature encoding on all images in the group of images to be identified to obtain a target histogram; and then identifies the target object based on the target histogram. In this technical solution, the group of images to be identified is obtained by performing two convolutions on the image to be processed, wherein the convolution kernel in the first convolution is constructed based on the image to be processed, and the convolution kernel in the second convolution is constructed based on a random image from multiple images generated by the first convolution. Furthermore, a binary hash coding process is performed on the group of images to be identified to obtain a target histogram, which highlights the structure and features of the target object based on the target histogram. The target histogram is then classified using a classifier to identify the target object. This solves the problem of inaccurate identification of the target object when the target object's face is obscured during facial recognition in the prior art. This solution achieves accurate identification of the target object even when the target object's face is obscured.

[0090] Optionally, the image group determination module includes: a first construction unit, configured to construct a first feature extraction operator based on the image to be processed;

[0091] a unit for determining a group of images to be used, configured to perform convolution processing on the image to be processed based on a first feature extraction operator to obtain a group of images to be used; wherein the group of images to be used includes at least one image to be used;

[0092] The image group to be identified determining unit is used to construct a second feature extraction operator based on the image group to be used, and perform convolution processing on each image to be used based on the second feature extraction operator to obtain the image group to be identified.

[0093] Optionally, the first construction unit includes: a first adjacency graph construction subunit, configured to construct a first adjacency graph based on the image to be processed; wherein the first adjacency graph includes at least one pixel node, and each pixel node has an adjacent edge with a pixel node within a preset neighboring range;

[0094] The first construction subunit is used to perform feature mapping according to each pixel node in the first adjacency graph and the adjacency edge weight corresponding to at least one adjacency edge to obtain a first feature extraction operator.

[0095] Optionally, the to-be-identified image group determination unit includes: a second adjacency graph construction subunit, configured to select a target use image from at least one to-be-used image and construct a second adjacency graph based on the target use image; wherein the second adjacency graph includes at least one pixel node, and each pixel node has an adjacent edge with a pixel node within a preset neighboring range;

[0096] The second construction subunit is used to perform feature mapping according to each pixel node in the second adjacency graph and the adjacent edge weight corresponding to at least one adjacent edge to obtain a second feature extraction operator.

[0097] Optionally, the object recognition module includes: an encoding unit, configured to perform binary hash coding processing on each image to be recognized in the image group to be recognized, to obtain a to-be-used histogram corresponding to each image to be recognized;

[0098] The target histogram determination unit is used to perform histogram combination on all histograms to be used to obtain a target histogram.

[0099] Optionally, the object recognition module includes: a classifier calling unit, configured to call a pre-built target classifier; wherein the target classifier is a classifier of an optimal path based on density measurement;

[0100] The object recognition unit is used to input the target histogram into the target classifier, obtain a classification result corresponding to the target object, and recognize the target object according to the classification result.

[0101] The target object recognition device provided by the embodiment of the present invention can execute the target object recognition method provided by any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0102] Example 4

[0103] Figure 5 A schematic diagram of the structure of an electronic device 10 of an embodiment of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or claimed herein.

[0104] like Figure 5As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., which is communicatively connected to the at least one processor 11. The memory stores a computer program that can be executed by the at least one processor. The processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 into the random access memory (RAM) 13. Various programs and data required for the operation of the electronic device 10 can also be stored in the RAM 13. The processor 11, ROM 12, and RAM 13 are connected to each other via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0105] Multiple components in the electronic device 10 are connected to the I / O interface 15, including an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a magnetic disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0106] The processor 11 can be any general-purpose and / or specialized processing component with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various specialized artificial intelligence (AI) computing chips, various processors that run machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as the target object recognition method.

[0107] In some embodiments, the target object recognition method can be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as the storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed on the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the target object recognition method described above can be performed. Alternatively, in other embodiments, the processor 11 can be configured to perform the target object recognition method in any other appropriate manner (for example, by means of firmware).

[0108] Various embodiments of the systems and techniques described above can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on a chip (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include being implemented in one or more computer programs that are executable and / or interpreted on a programmable system that includes at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0109] The computer program for implementing the target object identification method of the present invention can be written in any combination of one or more programming languages. These computer programs can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program can be executed entirely on the machine, partially on the machine, as a stand-alone software package, partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0110] In the context of the present invention, computer-readable storage media can be tangible media that can contain or store a computer program for use with an instruction execution system, device or equipment or used in combination with an instruction execution system, device or equipment. Computer-readable storage media can include but are not limited to electronic, magnetic, optical, electromagnetic, infrared or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, computer-readable storage media can be machine-readable signal media. More specific examples of machine-readable storage media can include electrical connections based on one or more lines, portable computer disks, hard disks, random access memories (RAM), read-only memories (ROM), erasable programmable read-only memories (EPROM or flash memory), optical fibers, portable compact disk read-only memories (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0111] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0112] The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0113] A computing system may include clients and servers. The clients and servers are typically remote from each other and typically interact via a communication network. This client-server relationship arises through computer programs running on the respective computers, creating a client-server relationship. The server may be a cloud server, also known as a cloud computing server or cloud host. This server is a hosting product within the cloud computing service ecosystem that addresses the management difficulties and limited scalability of traditional physical hosting and VPS services.

[0114] Example 5

[0115] An embodiment of the present invention further provides a computer program product, including a computer program, which, when executed by a processor, implements the target object recognition method provided in any embodiment of the present application.

[0116] The computer program product may be implemented by writing computer program code for performing the operations of the present invention 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 stand-alone 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 via 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., via the Internet using an Internet service provider).

[0117] It should be understood that the various forms of the processes shown above can be used to reorder, add, or delete steps. For example, the steps described in the present invention can be performed in parallel, sequentially, or in a different order, as long as the desired results of the technical solution of the present invention can be achieved. This is not limited herein.

[0118] The above specific embodiments do not limit the scope of protection of the present invention. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions may be made based on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention are intended to be included within the scope of protection of the present invention.

Claims

1. A method for identifying a target object, characterized in that: include: Obtain an image to be processed containing a target object; Performing multi-layer convolution processing on the image to be processed to obtain a group of images to be identified corresponding to the image to be processed; Perform feature coding processing on all images in the to-be-identified image group to obtain a target histogram, and perform object recognition on the target object based on the target histogram.

2. The method according to claim 1, characterized in that The step of performing multi-layer convolution processing on the image to be processed to obtain a group of images to be identified corresponding to the image to be processed includes: Constructing a first feature extraction operator based on the image to be processed; Performing convolution processing on the image to be processed based on the first feature extraction operator to obtain a group of images to be used; wherein the group of images to be used includes at least one image to be used; A second feature extraction operator is constructed based on the image group to be used, and convolution processing is performed on each image to be used based on the second feature extraction operator to obtain an image group to be identified.

3. The method according to claim 2, characterized in that The constructing a first feature extraction operator based on the image to be processed includes: Constructing a first adjacency graph based on the image to be processed; wherein the first adjacency graph includes at least one pixel node, and each pixel node has an adjacent edge with pixel nodes within a preset neighboring range; Feature mapping is performed according to each pixel node in the first adjacency graph and an adjacency edge weight corresponding to at least one adjacency edge to obtain a first feature extraction operator.

4. The method according to claim 2, characterized in that The constructing a second feature extraction operator based on the to-be-used image group includes: Selecting a target use image from at least one of the to-be-used images, and constructing a second adjacency graph based on the target use image; wherein the second adjacency graph includes at least one pixel node, and each pixel node has an adjacent edge with a pixel node within a preset neighboring range; Feature mapping is performed according to each pixel node in the second adjacency graph and an adjacent edge weight corresponding to at least one adjacent edge to obtain a second feature extraction operator.

5. The method according to claim 1, wherein The step of performing feature encoding processing on all images in the to-be-identified image group to obtain a target histogram includes: Performing binary hash coding processing on each image to be identified in the group of images to be identified, to obtain a to-be-used histogram corresponding to each image to be identified; Perform histogram combination on all histograms to be used to obtain the target histogram.

6. The method according to claim 1, characterized in that The performing object recognition on the target object based on the target histogram includes: Retrieving a pre-built target classifier; wherein the target classifier is a classifier of an optimal path based on density measurement; The target histogram is input into the target classifier to obtain a classification result corresponding to the target object, and the target object is identified according to the classification result.

7. A target object recognition device, characterized in that: include: An acquisition module, used for acquiring an image to be processed containing a target object; An image group determination module is used to perform multi-layer convolution processing on the image to be processed to obtain an image group to be identified corresponding to the image to be processed; The object recognition module is used to perform feature coding processing on all images in the image group to be recognized to obtain a target histogram, and perform object recognition on the target object based on the target histogram.

8. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor. The computer program is executed by the at least one processor to enable the at least one processor to perform the target object recognition method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the target object recognition method according to any one of claims 1 to 6 when executed.

10. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the computer program implements the target object recognition method according to any one of claims 1 to 6.

Images