Image classification method, device and electronic equipment

Through image classification technology, the similarity between images and standard image subclasses is calculated and the image subclass is determined, which solves the problem of low image classification accuracy in the prior art, and realizes the accurate and multi-dimensional classification results of images.

CN111191665BActive Publication Date: 2024-11-05BEIJING QIHOOD TECHNOLOGY CO LTD
View PDF 9 Cites 0 Cited by

Patent Information

Application Number
CN201811354310.8
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2018-11-14
Publication Date
2024-11-05
Estimated Expiration
2038-11-14

AI Technical Summary

Technical Problem

The existing image classification technology can only realize the preliminary classification of images and cannot determine the subclass to which images belong in this classification, resulting in low classification accuracy and affecting the efficiency of subsequent analysis.

Method used

By acquiring the image to be analyzed, image classification is performed, the similarity between the image and the standard image subclass is calculated, and the subclass with a similarity greater than a predetermined threshold is determined as an image subclass, and the comprehensive classification results are determined based on each image subclass.

Benefits of technology

The precise classification of images is achieved, the accuracy and multidimensionality of classification results are improved, and a detailed classification basis is provided for subsequent analysis.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN111191665B_ABST
    Figure CN111191665B_ABST
Patent Text Reader

Abstract

The embodiment of the present application discloses an image classification method, device and electronic device. The method includes: obtaining at least one image to be analyzed of the target object; classifying each image to be analyzed, and determining the image classification corresponding to any image to be analyzed; calculating the similarity between any image to be analyzed and the standard image corresponding to at least one image subclass corresponding to it; determining the image subclass with a similarity greater than a predetermined similarity threshold as the image subclass corresponding to any image to be analyzed; and determining the comprehensive classification result for the target object according to the image subclass corresponding to each image to be analyzed. The embodiment of the present application achieves the purpose of accurately classifying at least one image to be analyzed of the target object, provides a basis for the subsequent determination of the comprehensive classification result, and thus improves the accuracy of the comprehensive classification result.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of image recognition technology, and in particular to an image classification method, device and electronic equipment. Background Art

[0002] Image classification refers to the technology of using computers to classify an image or each pixel or area in an image into one of several categories to replace human visual interpretation. Image classification can improve the efficiency of subsequent analysis based on the classification results. However, the existing related image classification technology can only realize the preliminary classification of images, and cannot further determine the subclass to which the image belongs in the classification. For example, it can only be determined that the image to be analyzed is a peony, but it cannot be determined which type of peony it is. Summary of the invention

[0003] In view of this, the embodiments of the present application provide an image classification method, device and electronic device, which can achieve the above technical problems.

[0004] In order to solve the above problems, the embodiments of the present application mainly provide the following technical solutions:

[0005] In a first aspect, an embodiment of the present application provides an image classification method, the method comprising:

[0006] Acquire at least one image to be analyzed corresponding to the target;

[0007] Classifying each image to be analyzed, and determining an image classification corresponding to any image to be analyzed, wherein the image classification includes at least one image subclass;

[0008] Calculate the similarity between any image to be analyzed and the standard image corresponding to at least one image subclass;

[0009] Determine the image subclass whose similarity is greater than a predetermined similarity threshold as the image subclass corresponding to any image to be analyzed;

[0010] According to the image subclasses corresponding to each image to be analyzed, a comprehensive classification result for the target object is determined.

[0011] In a second aspect, an embodiment of the present application further provides an image classification device, the device comprising:

[0012] An image acquisition module, used to acquire at least one image to be analyzed of the target object;

[0013] An image classification module, used to classify each image to be analyzed, and determine the image classification corresponding to any image to be analyzed, wherein the image classification includes at least one image subclass;

[0014] A similarity calculation module, used to calculate the similarity between any image to be analyzed and a standard image corresponding to at least one image subclass corresponding to it;

[0015] A subclass determination module, used to determine an image subclass with a similarity greater than a predetermined similarity threshold as an image subclass corresponding to any image to be analyzed;

[0016] The comprehensive analysis module is used to determine the comprehensive classification result corresponding to the target according to the image subclasses corresponding to each image to be analyzed.

[0017] In a third aspect, an embodiment of the present application further provides an electronic device, including:

[0018] at least one processor;

[0019] and at least one memory and bus connected to the processor; wherein,

[0020] The processor and memory communicate with each other through the bus;

[0021] The processor is used to call the program instructions in the memory to execute the above-mentioned image classification method.

[0022] In a fourth aspect, an embodiment of the present application further provides a non-transitory computer-readable storage medium, which stores computer instructions, and the computer instructions enable a computer to execute the above-mentioned image classification method.

[0023] The technical solution provided in the embodiment of the present application has the following beneficial effects: obtaining at least one image to be analyzed of the target object, classifying each image to be analyzed, determining the image classification corresponding to any image to be analyzed, wherein the image classification includes at least one image subclass, calculating the similarity between any image to be analyzed and the standard image corresponding to at least one image subclass corresponding to it, and determining the image subclass with a similarity greater than a predetermined similarity threshold as the image subclass corresponding to any image to be analyzed, thereby achieving the purpose of accurately classifying at least one image to be analyzed of the target object and providing a basis for subsequently determining the comprehensive classification result; determining the comprehensive classification result for the target object according to the image subclasses corresponding to each image to be analyzed, and obtaining a multi-dimensional and comprehensive classification result with high accuracy for the target object. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in describing the embodiments of the present application are briefly introduced below.

[0025] Figure 1 A schematic diagram of a process flow of an image classification method provided in an embodiment of the present application is shown;

[0026] Figure 2A schematic diagram of a process flow of an image classification method provided in an embodiment of the present application is shown;

[0027] Figure 3 A schematic diagram of the structure of an image classification device provided in an embodiment of the present application is shown;

[0028] Figure 4 A schematic structural diagram of an electronic device provided in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0029] The embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present application, and cannot be interpreted as limiting the present application.

[0030] It will be understood by those skilled in the art that, unless expressly stated, the singular forms "one", "said", and "the" used herein may also include plural forms. It should be further understood that the term "comprising" used in the specification of the present application refers to the presence of the features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or groups thereof. It should be understood that when we refer to an element as being "connected" or "coupled" to another element, it may be directly connected or coupled to the other element, or there may be an intermediate element. In addition, the "connection" or "coupling" used herein may include wireless connection or wireless coupling. The term "and / or" used herein includes all or any unit and all combinations of one or more associated listed items.

[0031] In order to make the objectives, technical solutions and advantages of the present application clearer, the implementation methods of the present application will be further described in detail below with reference to the accompanying drawings.

[0032] In the existing technology, image classification can only determine the major category to which the image belongs, but cannot determine the minor category to which it belongs. For example, it can only determine that an image is a peony, but cannot determine which variety of peony it is. Therefore, the image classification accuracy is low, which directly affects the efficiency of subsequent analysis based on the classification results.

[0033] The image classification method, device, electronic device and computer-readable storage medium provided in this application are intended to solve the above technical problems in the prior art.

[0034] The technical solution of the present application and how the technical solution of the present application 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 application will be described below in conjunction with the accompanying drawings.

[0035] In one embodiment, Figure 1 As shown, the present application provides an image classification method, including: step S101 to step S105.

[0036] Step S101: Acquire at least one image to be analyzed of a target object.

[0037] In the embodiment of the present application, the target object can be determined according to actual needs. Specifically, the target object can be a person, an animal, a plant or other types.

[0038] In practical applications, at least one image of the target object to be analyzed can be an image of the same position of the target object or an image of different positions. For example, assuming that two images of the target object to be analyzed are obtained, if the target object is a person, one of the two images can be a tongue image and the other can be a whole face image.

[0039] In actual application, the image to be analyzed can be collected in real time by the current user using a terminal device with image acquisition function, or can be obtained from image data stored in a storage medium such as a database. Specifically, the terminal device can be an electronic device such as a mobile phone, a pad, a notebook, a wearable device, etc.

[0040] Step S102: classify each image to be analyzed, and determine an image classification corresponding to any image to be analyzed, where the image classification includes at least one image subclass.

[0041] In an embodiment of the present application, image classification is used to characterize the type (i.e., major category) corresponding to the target object. For example, if the target object refers to a user (i.e., a person. For the sake of convenience in explaining the present application, the following description will be based on the assumption that the target object refers to a human being), then the determination of the image classification realizes the preliminary classification of the image to be analyzed, thereby determining the scope of the image subclasses, achieving the purpose of narrowing the scope of the image subclasses, and providing a guarantee for subsequently improving the efficiency of the similarity calculation.

[0042] For example, if the image classification corresponding to the image to be analyzed points to the tongue, then the image subclass pointed to by the image classification tongue corresponds to the specific classification of the tongue, which can be specifically expressed as the image subclass of the tongue color, the image subclass of the tongue texture, etc.

[0043] Step S103: Calculate the similarity between any image to be analyzed and the standard image corresponding to at least one image subclass.

[0044] In the embodiment of the present application, the similarity is used to characterize the degree of similarity between any image to be analyzed and the standard image corresponding to at least one image subclass corresponding to it.

[0045] In practical application, it is assumed that there are six image classifications, namely image classification 1 to image classification 6, and it is assumed that two images to be analyzed are obtained, namely image 1 to be analyzed and image 2 to be analyzed, wherein image 1 to be analyzed corresponds to image classification 1, and image 2 to be analyzed corresponds to image classification 2. The subclasses of image classification 1 are subclass a and subclass b, and the subclass of image classification 2 is subclass c. Then, the similarities between image 1 to be analyzed and the standard image corresponding to subclass a and the standard image corresponding to subclass b are calculated respectively, as well as the similarity between image 2 to be analyzed and the standard image corresponding to subclass c.

[0046] In actual application, the histogram algorithm, perceptual hash algorithm, average hash algorithm, difference hash algorithm, scale-invariant feature algorithm, etc. can be used to calculate the similarity, which will not be described in detail in this application.

[0047] Step S104: Determine the image subclasses whose similarity is greater than a predetermined similarity threshold as the image subclasses corresponding to any image to be analyzed.

[0048] In actual application, the greater the similarity, the greater the similarity between the two images. The present application determines the division criteria for dividing the image subclasses by setting the similarity threshold.

[0049] For example, the image classification corresponding to the image 1 to be analyzed is the tongue (i.e., image classification), and the tongue includes three image subclasses, namely, subclass 1, subclass 2, and subclass 3. By calculating, the similarity between the image 1 to be analyzed and the standard image corresponding to subclass 1 is 70%, the similarity with the standard image corresponding to subclass 2 is 80%, and the similarity with the standard image corresponding to subclass 3 is 75%. If the predetermined similarity threshold is 70%, subclass 2 and subclass 3 are determined to be the image subclasses corresponding to the image 1 to be analyzed.

[0050] Step S105: Determine a comprehensive classification result corresponding to the target according to the image subclasses corresponding to the images to be analyzed.

[0051] The present application achieves the purpose of comprehensively determining the comprehensive classification result by performing comprehensive analysis on the image subclasses corresponding to each image to be analyzed, thereby improving the accuracy of the comprehensive classification result.

[0052] For example, the image subclasses determined in step S104 include three image subclasses: image subclass A, image subclass B, and image subclass C. The embodiment of the present application performs another analysis based on these three image subclasses to obtain a comprehensive classification result corresponding to the target object.

[0053] In an embodiment of the present application, at least one image to be analyzed of the target object is obtained, each image to be analyzed is classified, and an image classification corresponding to any image to be analyzed is determined, wherein the image classification includes at least one image subclass, and similarity between any image to be analyzed and a standard image corresponding to at least one image subclass corresponding to the image to be analyzed is calculated, and an image subclass with a similarity greater than a predetermined similarity threshold is determined as the image subclass corresponding to any image to be analyzed, thereby achieving the purpose of accurately classifying at least one image to be analyzed of the target image, and providing a basis for subsequently determining a comprehensive classification result; a comprehensive classification result for the target object is determined based on the image subclasses corresponding to each image to be analyzed, and a multi-dimensional and comprehensive classification result with high accuracy for the target object can be obtained.

[0054] In the embodiment of the present application, when humans are the target objects, image classification is divided according to the structure of the human body. For example, image classification can be divided into tongue, hand, eye, face, limbs and other categories.

[0055] In one implementation, in the embodiment of the present application, step S102 classifies each image to be analyzed and determines the image classification corresponding to any image to be analyzed, including: step S1021 and step S1022.

[0056] Step S1021 (not shown in the figure), input any image to be analyzed into a preset image recognition model to determine the corresponding image classification.

[0057] In actual application, the image recognition model can be a model constructed using a convolutional neural network. Before use, the constructed convolutional neural network is trained with a pre-collected sample set to obtain an image recognition model, so that the image recognition model can be used to achieve the purpose of automatically identifying the image classification corresponding to the image to be analyzed.

[0058] Step S1022 (not shown in the figure): determining the image category corresponding to any image to be analyzed according to the image category selection instruction input by the user for any image to be analyzed.

[0059] In actual application, selectable image classifications (such as image classification 1, image classification 2, image classification 3) can be configured on a preset interactive interface, and the user inputs the selected image classification through the interactive interface so that the image classification corresponding to any image to be analyzed can be determined according to the selection instruction input by the user.

[0060] It should be noted that two methods for determining image classification are provided in the embodiment of the present application. In actual application, only step S1021 or step S1022 may be referred to to provide a method for determining image classification.

[0061] In another implementation, Figure 1 As shown, step S103 includes: step S1031 and step S1032, wherein:

[0062] Step S1031 (not shown in the figure), extracting image features of any image to be analyzed based on a preset feature extraction model for image classification;

[0063] Step S1032 (not shown in the figure), calculating the similarity between the image features and the image features of the standard images corresponding to each image subclass corresponding to any image to be analyzed.

[0064] The embodiment of the present application ensures the accuracy of extracted image features, speeds up the extraction of image features, and ensures the accuracy of similarity calculation by using a feature extraction model for image classification.

[0065] In practical applications, the feature extraction model for different image classifications can be a CNN convolutional neural network model to achieve the purpose of multi-dimensional extraction of image features of any image to be analyzed. Specifically, the image features of the standard image corresponding to each image subclass can be pre-stored, so the image feature extraction step of the standard image corresponding to each image subclass is saved in the similarity calculation process, thereby achieving the purpose of simplifying the similarity calculation.

[0066] In another implementation, in an embodiment of the present application, step S103 calculates the similarity between any image to be analyzed and the standard image corresponding to at least one image subclass corresponding to it, including: when the number of standard images corresponding to any image subclass exceeds a predetermined number threshold, calculating the average value of the similarities between the corresponding image to be analyzed and the respective standard images corresponding to any image subclass, and using the average value as the similarity between the image to be analyzed and the standard image corresponding to any image subclass.

[0067] In practical application, if the image to be analyzed is image X, and if it is assumed that the number of standard images corresponding to image subclass a is 1 (i.e., the predetermined number threshold), then there is only one result for the similarity between image X and the standard image corresponding to image subclass a, and this similarity is the similarity between image X and the standard image corresponding to image subclass a; assuming that the number of standard images corresponding to image subclass a is 3, then there are three results for the similarity between image X and the standard image corresponding to image subclass a. At this time, it is necessary to take the average of these three similarities as the similarity between image X and the standard image corresponding to image subclass a.

[0068] In yet another implementation, Figure 1 As shown, the method also includes:

[0069] Step S106 (not shown in the figure), obtaining the priority information of the image classification corresponding to each determined image to be analyzed;

[0070] Step S105 determines a comprehensive classification result for the target object according to the image subclasses corresponding to each image to be analyzed, including: determining a comprehensive classification result for the target object based on priority information of the image classification corresponding to each image to be analyzed and according to the image subclasses corresponding to each image to be analyzed.

[0071] In the embodiment of the present application, the priority information is used to characterize the importance of determining the comprehensive classification result.

[0072] In practical applications, if the image classifications corresponding to any image to be analyzed determined in step S102 include classification 1, classification 2, and classification 3, and if the image subclasses corresponding to these three classifications are the same, they all include subclasses 1 to 5. Assume that the image subclass determined according to classification 1 is subclass 2, the image subclass determined according to classification 3 is subclass 3, and the image subclass determined according to classification 3 is classification 4, then the comprehensive classification result can be determined according to the priority information, and assuming that the priority of classification 1 is higher, then the final comprehensive classification result is subclass 2.

[0073] In another implementation, step S105 determines a comprehensive classification result for the target object according to the image subclasses corresponding to the images to be analyzed, including:

[0074] According to the image subclasses corresponding to the images to be analyzed, and in combination with the object-related information of the target object, a comprehensive classification result for the target object is determined.

[0075] Specifically, the object-related information includes at least one of the following: object attribute information, and environment information corresponding to the target object.

[0076] In actual application, the object attribute information can be determined according to the type of the target object. For example, when the target object is a human being, the object attribute information can be the age, gender, occupation, etc. of the target object; and the environmental information corresponding to the target object can be factors such as the living area (such as Beijing) and climate.

[0077] The embodiment of the present application uses the object-related information of the target object as a consideration for determining the comprehensive classification result, which provides a necessary basis for subsequent analysis of the causes and provision of suggestions after the comprehensive classification result is determined.

[0078] In yet another embodiment, Figure 2As shown, the method includes: step S201 to step S206, wherein step S201, step S202, step S203, step S204, step S205 are respectively the same or similar to the execution process of step S101, step S102, step S103, step S104, step S105, and are not repeated here.

[0079] In step S206, if the comprehensive classification result is a predetermined result and the determined image classification does not include the predetermined image classification, a prompt message is generated to obtain at least one image to be analyzed of the predetermined image classification, so as to re-determine the comprehensive classification result.

[0080] For example, assuming that the comprehensive classification result is determined to be an image subclass of abnormal tongue coating color, if the image classification determined in step S202 does not include facial image classification, then the image subclass determined to be abnormal tongue coating color may be inaccurate. In order to improve the accuracy of the comprehensive classification result, it is necessary to obtain the image to be analyzed for the facial image classification (i.e., the predetermined image classification) in order to re-determine the comprehensive classification result.

[0081] In one embodiment, Figure 3 As shown, a schematic diagram of the structure of an image classification device provided in an embodiment of the present application is shown in FIG. Figure 3 As shown, the device 40 includes:

[0082] An image acquisition module 401 is used to acquire at least one image to be analyzed of a target object;

[0083] An image classification module 402 is used to classify each image to be analyzed and determine an image classification corresponding to any image to be analyzed, wherein the image classification includes at least one image subclass;

[0084] A similarity calculation module 403 is used to calculate the similarity between any image to be analyzed and a standard image corresponding to at least one image subclass corresponding to it;

[0085] A subclass determination module 404, configured to determine an image subclass with a similarity greater than a predetermined similarity threshold as an image subclass corresponding to any image to be analyzed;

[0086] The comprehensive analysis module 405 is used to determine the comprehensive classification result corresponding to the target according to the image subclasses corresponding to each image to be analyzed.

[0087] The image classification device provided by the present application obtains at least one image to be analyzed of the target object, classifies each image to be analyzed, determines the image classification corresponding to any image to be analyzed, and the image classification includes at least one image subclass. The similarity between any image to be analyzed and the standard image corresponding to at least one image subclass corresponding to it is calculated, and the image subclass with a similarity greater than a predetermined similarity threshold is determined as the image subclass corresponding to any image to be analyzed, thereby achieving the purpose of accurately classifying at least one image to be analyzed of the target image, and providing a basis for subsequently determining the comprehensive classification result; according to the image subclasses corresponding to each image to be analyzed, the comprehensive classification result for the target object is determined, and a multi-dimensional and comprehensive classification result with high accuracy for the target object can be obtained.

[0088] Furthermore, when humans are the target objects, image classification is divided according to the structure of the human body.

[0089] Furthermore, the image classification module 402 is used to input any image to be analyzed into a preset image recognition model to determine the corresponding image classification; or to determine the image classification corresponding to any image to be analyzed according to the image classification selection instruction received from the user for any image to be analyzed.

[0090] Furthermore, the similarity calculation module 403 is also used to extract image features of any image to be analyzed based on a preset feature extraction model for image classification; and calculate the similarity between the image features and image features of standard images corresponding to each image subclass corresponding to any image to be analyzed.

[0091] Furthermore, the similarity calculation module 403 is also used to calculate the average value of the similarities between the corresponding image to be analyzed and the standard images corresponding to any image subclass when the number of standard images corresponding to any image subclass exceeds a predetermined threshold value, and use the average value as the similarity between the image to be analyzed and the standard images corresponding to any image subclass.

[0092] Furthermore, the comprehensive analysis module 405 is also used to obtain the priority information of the image classification corresponding to each image to be analyzed; based on the priority information of each image to be analyzed and according to the image subclasses corresponding to each image to be analyzed, determine the comprehensive classification result for the target object.

[0093] Furthermore, the comprehensive analysis module 405 is further used to determine a comprehensive classification result for the target object according to the image subclasses corresponding to the images to be analyzed and in combination with the object-related information of the target object.

[0094] Furthermore, the comprehensive analysis module 405 is also used to generate prompt information for obtaining at least one image to be analyzed of the predetermined image classification if the comprehensive classification result is a predetermined result and the predetermined image classification is not included in the determined image classification, so as to re-determine the comprehensive classification result.

[0095] Furthermore, the object-related information includes at least one of the following: object attribute information, and environment information corresponding to the target object.

[0096] The image classification device of this embodiment can execute the image classification method provided in the embodiment of the present application. The implementation principle is similar and will not be repeated here.

[0097] In one embodiment, the present application provides an electronic device, such as Figure 4 As shown, Figure 4 The electronic device 600 shown includes: a processor 6001 and a memory 6003. The processor 6001 and the memory 6003 are connected, such as through a bus 6002. Further, the electronic device 600 may also include a transceiver 6006, wherein the transceiver is used to provide the electronic device with a communication service function between other devices. It should be noted that in actual applications, the transceiver 6006 is not limited to one, and the structure of the electronic device 600 does not constitute a limitation on the embodiments of the present application.

[0098] The processor 6001 may be a CPU, a general-purpose processor, a DSP, an ASIC, an FPGA or other programmable logic device, a transistor logic device, a hardware component 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. The processor 6001 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.

[0099] The bus 6002 may include a path to transmit information between the above components. The bus 6002 may be a PCI bus or an EISA bus, etc. The bus 6002 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 4 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0100] The memory 6003 may be a ROM or other type of static storage device that can store static information and instructions, a RAM or other type of dynamic storage device that can store information and instructions, or an EEPROM, a CD-ROM or other optical disk storage, an optical disk storage (including a compressed optical disk, a laser disk, an optical disk, a digital versatile disk, a 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 these.

[0101] The memory 6003 is used to store the application code for executing the solution of the present application, and the execution is controlled by the processor 6001. The processor 6001 is used to execute the application code stored in the memory 6003 to implement Figure 3 The illustrated embodiment provides an image classification device.

[0102] The electronic device provided in the embodiment of the present application obtains at least one image to be analyzed of the target object, classifies each image to be analyzed, determines the image classification corresponding to any image to be analyzed, the image classification includes at least one image subclass, calculates the similarity between any image to be analyzed and the standard image corresponding to at least one image subclass corresponding to it, and determines the image subclass with a similarity greater than a predetermined similarity threshold as the image subclass corresponding to any image to be analyzed, thereby achieving the purpose of accurately classifying at least one image to be analyzed of the target image and providing a basis for subsequently determining the comprehensive classification result; according to the image subclasses corresponding to each image to be analyzed, the comprehensive classification result for the target object is determined, and a multi-dimensional and comprehensive classification result with high accuracy for the target object can be obtained.

[0103] In one embodiment, the present application provides a non-transitory computer-readable storage medium, which stores computer instructions, and the computer instructions enable a computer to execute the above-mentioned image classification method.

[0104] A non-transitory computer-readable storage medium provided in an embodiment of the present application, compared with the prior art, obtains at least one image to be analyzed of a target object, classifies each image to be analyzed, determines an image classification corresponding to any image to be analyzed, the image classification includes at least one image subclass, calculates the similarity between any image to be analyzed and a standard image corresponding to at least one image subclass corresponding to it, and determines an image subclass with a similarity greater than a predetermined similarity threshold as an image subclass corresponding to any image to be analyzed, thereby achieving the purpose of accurately classifying at least one image to be analyzed of the target image, and providing a basis for subsequently determining a comprehensive classification result; according to the image subclasses corresponding to each image to be analyzed, a comprehensive classification result for the target object is determined, and a multi-dimensional and comprehensive classification result with high accuracy for the target object can be obtained.

[0105] It should be understood that, although the steps in the flowchart of the accompanying drawings are displayed in sequence as indicated by the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least a part of the steps in the flowchart of the accompanying drawings may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be executed in turn or alternately with other steps or at least a part of the sub-steps or stages of other steps.

[0106] The above are only some implementation methods of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. An image classification method, characterized in that: include: Acquire at least one image to be analyzed of the target object; Classifying each image to be analyzed, and determining an image classification corresponding to any image to be analyzed, wherein the image classification includes at least one image subclass, the image classification is used to characterize a partial position corresponding to a target object, and each image subclass is a partial feature of the image classification; Calculating the similarity between any of the images to be analyzed and the standard images corresponding to at least one image subclass corresponding to it; Determine the image subclass whose similarity is greater than a predetermined similarity threshold as the image subclass corresponding to any one of the images to be analyzed; A comprehensive classification result for the target object is determined according to the image subclasses corresponding to the images to be analyzed.

2. The method according to claim 1, characterized in that The step of classifying each image to be analyzed and determining an image classification corresponding to any image to be analyzed includes: Input any image to be analyzed into a preset image recognition model to determine the corresponding image classification; or According to the received image classification selection instruction input by the user for any image to be analyzed, the image classification corresponding to the any image to be analyzed is determined.

3. The method according to claim 1, characterized in that The calculating the similarity between any of the images to be analyzed and the standard images corresponding to at least one of the corresponding image subclasses includes: Extracting image features of any image to be analyzed based on a preset feature extraction model for the image classification; The similarity between the image feature and the image feature of the standard image corresponding to each image subclass corresponding to any one of the images to be analyzed is calculated.

4. The method according to claim 1, characterized in that The calculating the similarity between any of the images to be analyzed and the standard images corresponding to at least one of the corresponding image subclasses includes: When the number of standard images corresponding to any image subclass exceeds a predetermined threshold, the average value of the similarities between the corresponding image to be analyzed and each standard image corresponding to any image subclass is calculated, and the average value is used as the similarity between the image to be analyzed and the standard image corresponding to any image subclass.

5. The method according to claim 1, characterized in that The method further includes: Obtaining priority information of image classifications corresponding to each determined image to be analyzed; Determining the comprehensive classification result for the target object according to the image subclasses corresponding to the images to be analyzed includes: Based on the priority information of each image to be analyzed and according to the image subclasses respectively corresponding to each image to be analyzed, a comprehensive classification result for the target object is determined.

6. The method according to claim 1, characterized in that Determining the comprehensive classification result for the target object according to the image subclasses corresponding to the images to be analyzed includes: A comprehensive classification result for the target object is determined according to the image subclasses corresponding to the images to be analyzed and in combination with the object-related information of the target object.

7. The method according to claim 6, characterized in that The object related information includes at least one of the following: Object attribute information; environment information corresponding to the target object.

8. The method according to claim 1, characterized in that When humans are the target objects, the image classification is divided according to the structure of the human body.

9. The method according to claim 1, characterized in that: The method further comprises: If the comprehensive classification result is a predetermined result, and the determined image classification does not include the predetermined image classification, prompt information for obtaining at least one image to be analyzed of the predetermined image classification is generated to redetermine the comprehensive classification result.

10. An image classification device, characterized in that: include: An image acquisition module, used to acquire at least one image to be analyzed of the target object; An image classification module is used to classify each image to be analyzed and determine the image classification corresponding to any image to be analyzed, wherein the image classification includes at least one image subclass, and the image classification is used to characterize a partial position corresponding to a target object, and each image subclass is a partial feature of the image classification; A similarity calculation module, used for calculating the similarity between any image to be analyzed and a standard image corresponding to at least one image subclass corresponding thereto; A subclass determination module, configured to determine an image subclass with a similarity greater than a predetermined similarity threshold as an image subclass corresponding to any image to be analyzed; The comprehensive analysis module is used to determine the comprehensive classification result for the target object according to the image subclasses corresponding to each image to be analyzed.

11. An electronic device, characterized in that: include: at least one processor; and at least one memory and bus connected to the processor; wherein, The processor and the memory communicate with each other via the bus; The processor is used to call the program instructions in the memory to execute the image classification method according to any one of claims 1 to 9.

12. A non-transitory computer-readable storage medium, characterized in that: The non-transitory computer-readable storage medium stores computer instructions, and the computer instructions cause the computer to execute the image classification method according to any one of claims 1 to 9.

Citation Information

Patent Citations

  • Road sign detection method and device

    CN103034836A

  • Fine granularity classification recognition method and object part location and feature extraction method thereof

    CN104573744A

  • Application uninstalling reason classification method and classification apparatus

    CN104933044A

  • Metal classification image display method and device

    CN105911602A

  • Clustering and reclassifying face recognition method

    CN106250821A