Image recognition method, device, computer equipment and storage medium

Abstract

The present disclosure proposes an image recognition method, apparatus, computer equipment and storage medium. The method includes: acquiring an image to be recognized, processing the image to be recognized, obtaining grayscale description information, determining an image recognition sequence based on the grayscale description information, and performing target recognition on the image to be recognized based on the grayscale description information and the image recognition sequence. Thus, by acquiring the grayscale description information and the image recognition sequence of the image to be recognized to perform image recognition, a reliable reference basis can be provided for the target recognition process of the image to be recognized, thereby effectively improving the target recognition effect of the image to be recognized.

Description

Technical Field

[0001] The present disclosure relates to the field of image processing technology, and in particular to an image recognition method, apparatus, computer equipment, and storage medium. Background Art

[0002] In the field of image processing technology, it is usually based on the risk relationship network, and the users associated with the known objects are identified based on their images. Then, the images of the associated users are integrated and analyzed to identify potential risks.

[0003] This approach does not work well for image recognition of known objects. Summary of the Invention

[0004] The present disclosure aims to solve one of the technical problems in the related art at least to a certain extent.

[0005] To this end, the purpose of the present disclosure is to propose an image recognition method, device, computer equipment and storage medium, which can provide a reliable reference basis for the target recognition process of the image to be recognized by obtaining the grayscale description information and image recognition sequence of the image to be recognized, thereby effectively improving the target recognition effect of the image to be recognized.

[0006] The image recognition method proposed in the embodiment of the first aspect of the present disclosure includes: acquiring an image to be recognized; processing the image to be recognized to obtain grayscale description information; determining an image recognition sequence based on the grayscale description information; and performing target recognition on the image to be recognized based on the grayscale description information and the image recognition sequence.

[0007] The image recognition method proposed in the embodiment of the first aspect of the present disclosure obtains the image to be recognized, processes the image to be recognized, obtains grayscale description information, determines the image recognition sequence based on the grayscale description information, and performs target recognition on the image to be recognized based on the grayscale description information and the image recognition sequence. Therefore, by obtaining the grayscale description information and the image recognition sequence of the image to be recognized to perform image recognition, a reliable reference basis can be provided for the target recognition process of the image to be recognized, thereby effectively improving the target recognition effect of the image to be recognized.

[0008] The image recognition device proposed in the second aspect embodiment of the present disclosure includes: an acquisition module for acquiring an image to be recognized; a processing module for processing the image to be recognized to obtain grayscale description information; a determination module for determining an image recognition sequence based on the grayscale description information; and an identification module for performing target recognition on the image to be recognized based on the grayscale description information and the image recognition sequence.

[0009] The image recognition device proposed in the embodiment of the second aspect of the present disclosure obtains the image to be recognized, processes the image to be recognized, obtains grayscale description information, determines the image recognition sequence based on the grayscale description information, and performs target recognition on the image to be recognized based on the grayscale description information and the image recognition sequence. Therefore, by obtaining the grayscale description information and the image recognition sequence of the image to be recognized to perform image recognition, a reliable reference basis can be provided for the target recognition process of the image to be recognized, thereby effectively improving the target recognition effect of the image to be recognized.

[0010] The computer device proposed in the third embodiment of the present disclosure includes: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the image recognition method proposed in the first embodiment of the present disclosure is implemented.

[0011] The fourth embodiment of the present disclosure proposes a non-temporary computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, it implements the image recognition method proposed in the first embodiment of the present disclosure.

[0012] The fifth embodiment of the present disclosure proposes a computer program product. When the instructions in the computer program product are executed by a processor, the image recognition method proposed in the first embodiment of the present disclosure is executed.

[0013] Additional aspects and advantages of the present disclosure will be given in part in the following description and in part will be obvious from the following description, or will be learned through practice of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] The above and / or additional aspects and advantages of the present disclosure will become apparent and readily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0015] Figure 1 is a flowchart of an image recognition method proposed in one embodiment of the present disclosure;

[0016] Figure 2 is a flowchart of an image recognition method proposed in another embodiment of the present disclosure;

[0017] Figure 3 is a schematic diagram of a process for processing an image to be recognized in an embodiment of the present disclosure;

[0018] Figure 4 is a flowchart of an image recognition method proposed in another embodiment of the present disclosure;

[0019] Figure 5 This is a schematic diagram of a target recognition process proposed in an embodiment of the present disclosure;

[0020] Figure 6is a flowchart of an image recognition method proposed in another embodiment of the present disclosure;

[0021] Figure 7 is a schematic diagram of the process of performing target recognition on an image to be recognized in an embodiment of the present disclosure;

[0022] Figure 8 is a structural diagram of an image recognition device proposed in one embodiment of the present disclosure;

[0023] Figure 9 is a structural diagram of an image recognition device proposed in another embodiment of the present disclosure;

[0024] Figure 10 A block diagram of an exemplary computer device suitable for implementing embodiments of the present disclosure is shown. DETAILED DESCRIPTION

[0025] The following describes in detail embodiments of the present disclosure, examples of which 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 intended only to explain the present disclosure and are not to be construed as limiting the present disclosure. On the contrary, the embodiments of the present disclosure include all variations, modifications, and equivalents that fall within the spirit and scope of the appended claims.

[0026] Figure 1 Schematic diagram of the image recognition method according to an embodiment of the present invention.

[0027] It should be noted that the executor of the image recognition method of this embodiment is an image recognition device, which can be implemented by software and / or hardware. The device can be configured in a computer device, which can include but is not limited to a terminal, a server, etc. For example, the terminal can be a mobile phone, a handheld computer, etc.

[0028] like Figure 1 As shown, the image recognition method includes:

[0029] S101: Acquire an image to be recognized.

[0030] The image to be recognized refers to the image to be used for target recognition, and the image to be recognized may be stored in the form of unstructured data. For example, it may be a qualification image used to register a company or store.

[0031] It is understandable that in the field of risk identification, the recognition efficiency and recognition effect of unstructured data are relatively poor. Therefore, in the embodiment of the present disclosure, by obtaining the image to be identified, the subsequent steps can be triggered in a timely manner to obtain the structured data corresponding to the image to be identified, such as grayscale description information and image recognition sequence, and the grayscale description information and image recognition sequence are used as the basis for identification to achieve target recognition of the image to be identified.

[0032] In the embodiment of the present disclosure, when obtaining the image to be identified, the user can upload the image to be identified to the execution entity of the embodiment of the present disclosure, or a database consisting of multiple images to be identified can be obtained in advance, and then the execution entity of the embodiment of the present disclosure can perform image recognition operations on each image to be identified in the database one by one. There is no limitation on this.

[0033] It should be noted that the images to be identified obtained in the embodiments of the present disclosure are all obtained after authorization by the relevant users, and the acquisition process complies with the provisions of relevant laws and regulations and does not violate public order and good morals.

[0034] S102: Process the image to be identified to obtain grayscale description information.

[0035] Grayscale refers to the logarithmic division between white and black in a grayscale image. Grayscale description information can be used to describe relevant information about the grayscale image to be identified, such as the grayscale distribution characteristics and grayscale extremes of the image to be identified.

[0036] In the embodiment of the present disclosure, the image to be identified can be grayscaled to obtain grayscale description information. The grayscale processing method can be based on component method, maximum method, average method, weighted averaging method, etc., and there is no limitation on this.

[0037] In the embodiment of the present disclosure, when processing the image to be identified and obtaining grayscale description information, feature enhancement processing can be performed on the image to be processed, and grayscale processing can be performed on the image after feature enhancement processing to obtain grayscale description information. Alternatively, encoding and compression processing can be performed on the image to be identified, and grayscale processing can be performed on the image to be processed after encoding and compression processing to obtain grayscale description information. There is no limitation on this.

[0038] It is understandable that the image to be recognized may be a color image. There are three components R, G, and B in the color image, which control the three basic colors of red, green, and blue respectively. The information provided by RGB is of low value and cannot reflect the morphological characteristics of the image, which may affect the work efficiency of the image recognition process.

[0039] Therefore, in the embodiment of the present disclosure, by processing the image to be identified and obtaining grayscale description information, it is possible to effectively reduce redundant information in the grayscale description information while ensuring the description effect of the obtained grayscale description information on the image to be identified, thereby providing a reliable reference basis for the determination process of the image recognition sequence and effectively improving the applicability of the obtained grayscale description information in the subsequent target recognition process.

[0040] S103: Determine an image recognition sequence according to the grayscale description information.

[0041] The image recognition sequence may refer to sequence information that maps the grayscale distribution features of the image to be recognized. For example, the grayscale description information may be processed using a hash algorithm to obtain a hash code corresponding to the image to be processed as the image recognition sequence.

[0042] It is understandable that the image recognition sequence can effectively represent the grayscale distribution information of the corresponding image to be recognized, and different images represent different features of objects, and their grayscale distribution information is also likely to be different.

[0043] In the embodiments of the present disclosure, when determining the image recognition sequence based on the grayscale description information, the grayscale description information can be input into a pre-trained machine learning model to obtain the corresponding image recognition sequence. Alternatively, any other possible method can be used to determine the image recognition sequence based on the grayscale description information, such as engineering or mathematical methods, and there is no limitation to this.

[0044] In the embodiment of the present disclosure, determining the image recognition sequence based on the grayscale description information can enable the obtained image recognition sequence to effectively characterize the grayscale distribution characteristics of the image to be recognized, and there is a high probability that there are differences between the image recognition sequences corresponding to different images to be recognized. Therefore, the image recognition sequence can provide a reliable reference basis for the subsequent target recognition process.

[0045] S104: Perform target recognition on the image to be recognized based on the grayscale description information and the image recognition sequence.

[0046] Among them, target recognition refers to the recognition process of the image to be recognized based on the preset recognition intention, grayscale description information and image recognition sequence.

[0047] In some embodiments, there are multiple images to be identified. When performing target recognition on the images to be identified based on the grayscale description information and the image recognition sequence, the multiple images to be identified can be grouped based on the grayscale description information and the image recognition sequence, so that the images to be identified in the same group have partially the same grayscale description information and / or image recognition sequence, so as to facilitate the retrieval of similar pictures.

[0048] In other embodiments, there are multiple images to be identified. When performing target recognition on the images to be identified based on the grayscale description information and the image recognition sequence, it can also be based on a pre-trained machine learning model to achieve target recognition of the images to be identified based on the grayscale description information and the image recognition sequence.

[0049] Alternatively, when performing target recognition on the image to be recognized based on the grayscale description information and the image recognition sequence, any other possible method, such as an engineering or mathematical method, may be used, and there is no limitation to this.

[0050] It can be understood that the grayscale description information and image recognition sequence belong to structured data, and there is a high probability that the grayscale description information and image recognition sequences of different images to be identified are different. Therefore, when target recognition is performed on the images to be identified based on the grayscale description information and image recognition sequence, the reliability of image recognition can be effectively improved and the image recognition effect can be improved.

[0051] In this embodiment, by acquiring the image to be identified and processing the image to be identified, grayscale description information is obtained, and based on the grayscale description information, an image recognition sequence is determined, and based on the grayscale description information and the image recognition sequence, target recognition is performed on the image to be identified. Thus, by acquiring the grayscale description information and the image recognition sequence of the image to be identified to perform image recognition, a reliable reference basis can be provided for the target recognition process of the image to be identified, thereby effectively improving the target recognition effect of the image to be identified.

[0052] Figure 2 It is a flowchart of an image recognition method proposed in another embodiment of the present disclosure.

[0053] like Figure 2 As shown, the image recognition method includes:

[0054] S201: Acquire an image to be recognized.

[0055] The description of S201 can be found in the above embodiment and will not be repeated here.

[0056] S202: Segment the image to be recognized to obtain multiple sub-images.

[0057] The sub-images refer to multiple images obtained by segmenting the image to be identified.

[0058] In the embodiment of the present disclosure, image segmentation of the image to be identified may be performed using a threshold-based segmentation method, a region-based segmentation method, an edge-based segmentation method, a segmentation method based on a specific theory, etc., and there is no limitation to this.

[0059] S203: Determine a plurality of grayscale information corresponding to the plurality of sub-images respectively.

[0060] The grayscale information refers to the relevant information describing the grayscale image corresponding to the sub-image, for example, it can be the grayscale values ​​corresponding to multiple sub-images.

[0061] In the embodiment of the present disclosure, grayscale processing can be performed on the image to be processed before image segmentation is performed on the image to be identified, and then image segmentation can be performed on the image to be processed after grayscale processing to determine multiple grayscale information corresponding to multiple sub-images respectively. Alternatively, grayscale processing can be performed on multiple sub-images after image segmentation is performed on the image to be identified to determine multiple grayscale information corresponding to multiple sub-images respectively. There is no limitation on this.

[0062] For example, if Figure 3 As shown, Figure 3 FIG. 1 is a flow chart of processing an image to be recognized in an embodiment of the present disclosure, wherein the flow may include:

[0063] Perform grayscale processing on the image to be identified to obtain a corresponding grayscale image;

[0064] Divide the obtained grayscale image into multiple 8*8 sub-images;

[0065] Using the grayscale information of each sub-image to describe the corresponding sub-image, the grayscale information may be, for example, the average grayscale value within the sub-image;

[0066] The grayscale information of each sub-image (such as the grayscale average) is displayed in an 8*8 table, which can also be used as an image recognition sequence.

[0067] S204: Generate grayscale description information according to the plurality of grayscale information.

[0068] In the embodiment of the present disclosure, when generating grayscale description information based on multiple grayscale information, it is possible to determine the grayscale median, grayscale extreme value, grayscale value mode and other related information of the multiple grayscale information, and select any combination thereof as the grayscale description information, without any limitation.

[0069] In the embodiment of the present disclosure, since multiple grayscale information can respectively represent the grayscale features of different sub-images, when grayscale description information is generated based on multiple grayscale information, the obtained grayscale description information can effectively represent the grayscale information distribution characteristics of the image to be identified, thereby providing a reliable reference object for the target recognition process.

[0070] That is to say, after obtaining the image to be identified, the embodiment of the present disclosure can perform image segmentation on the image to be identified to obtain multiple sub-images, and determine multiple grayscale information corresponding to the multiple sub-images, and then generate grayscale description information based on the multiple grayscale information. Since the grayscale features of different parts of the image to be identified may be different, when the image to be identified is segmented to obtain multiple sub-images and multiple grayscale information corresponding to the multiple sub-images are determined, the obtained multiple grayscale information can effectively characterize the grayscale features of the corresponding sub-images, and then generate grayscale description information based on the multiple grayscale information. The grayscale description information can clearly describe the distribution of the overall grayscale features of the image to be identified, thereby providing a reliable analysis object for the image recognition process.

[0071] S205: Obtain a preset grayscale threshold.

[0072] The preset grayscale threshold refers to a grayscale threshold value configured in advance for multiple grayscale information.

[0073] It is understandable that the grayscale distribution characteristics of the image to be processed may be uneven, and thus, the grayscale description information corresponding to multiple sub-images may also be different. In the embodiment of the present disclosure, by obtaining the preset grayscale threshold, a reliable comparison basis can be provided for the subsequent determination of multiple comparison results.

[0074] In the embodiment of the present disclosure, the preset grayscale thresholds corresponding to multiple sub-images can be the same, or any other possible method can be used to configure the corresponding preset grayscale thresholds for multiple sub-images. For example, the corresponding preset grayscale thresholds can be configured in combination with the relative positions of the sub-images, and there is no limitation on this.

[0075] S206: Compare the plurality of grayscale information with the preset grayscale thresholds respectively to obtain a plurality of comparison results.

[0076] The comparison result refers to the relevant information obtained after the grayscale information is compared with the preset grayscale threshold.

[0077] In the embodiment of the present disclosure, when multiple grayscale information are compared with the preset grayscale threshold respectively, the obtained multiple comparison results can be combined with the grayscale information and the preset grayscale threshold to effectively characterize the grayscale value characteristics of each sub-image.

[0078] S207: forming an image recognition sequence according to the plurality of comparison results and the plurality of sub-image identifiers corresponding thereto, wherein the sub-image identifier is used to identify the corresponding sub-image.

[0079] The sub-image identification refers to identification information corresponding to a sub-image, which can be used to identify multiple sub-images.

[0080] For example, in an embodiment of the present disclosure, corresponding serial number signals may be configured for multiple sub-images of an image to be processed to facilitate distinguishing and locating the multiple sub-images. The serial number information may serve as a sub-image identifier.

[0081] In the embodiment of the present disclosure, when forming an image recognition sequence based on multiple comparison results and multiple sub-image identifiers corresponding thereto, the difference information between multiple grayscale information and a preset grayscale threshold can be obtained, and the image recognition sequence can be generated based on the difference information and the corresponding sub-image identifier. Alternatively, the multiple comparison results and the multiple sub-image identifiers corresponding thereto can be input into a pre-trained machine learning model to obtain the corresponding image recognition sequence, and there is no limitation on this.

[0082] That is to say, after the grayscale description information, the embodiment of the present disclosure can obtain a preset grayscale threshold, compare multiple grayscale information with the preset grayscale threshold respectively, obtain multiple comparison results, and form an image recognition sequence based on the multiple comparison results and the multiple sub-image identifiers corresponding thereto. In this way, a preliminary analysis of the grayscale features of multiple sub-images in the image to be identified can be achieved, and the representation granularity of the obtained image recognition sequence can be improved, so that the image recognition sequence can accurately and intuitively represent the grayscale feature analysis results of each sub-image, thereby providing an accurate reference basis for subsequent target recognition.

[0083] S208: Perform target recognition on the image to be recognized based on the grayscale description information and the image recognition sequence.

[0084] The description of S208 can be found in the above embodiment and will not be repeated here.

[0085] In this embodiment, the image to be identified is segmented to obtain multiple sub-images, and multiple grayscale information corresponding to the multiple sub-images are determined. Then, grayscale description information is generated based on the multiple grayscale information. Since the grayscale features of different parts of the image to be identified may be different, when the image to be identified is segmented to obtain multiple sub-images, and multiple grayscale information corresponding to the multiple sub-images are determined, the obtained multiple grayscale information can effectively characterize the grayscale features of the corresponding sub-images. Then, grayscale description information is generated based on the multiple grayscale information. The grayscale description information can clearly describe the distribution of the overall grayscale features of the image to be identified, thereby providing a reliable analysis object for the image recognition process. A preset grayscale threshold is obtained, and multiple grayscale information are compared with the preset grayscale threshold respectively to obtain multiple comparison results. An image recognition sequence is formed based on the multiple comparison results and the multiple sub-image identifiers corresponding thereto. In this way, a preliminary analysis of the grayscale features of multiple sub-images in the image to be identified can be achieved, and the representation granularity of the obtained image recognition sequence can be improved, so that the image recognition sequence can accurately and intuitively represent the grayscale feature analysis results of each sub-image, thereby providing an accurate reference basis for subsequent target recognition.

[0086] Figure 4 It is a flowchart of an image recognition method proposed in another embodiment of the present disclosure.

[0087] like Figure 4 As shown, the image recognition method includes:

[0088] S401: Acquire an image to be recognized.

[0089] S402: Segment the image to be recognized to obtain multiple sub-images.

[0090] S403: Determine a plurality of grayscale information corresponding to the plurality of sub-images respectively.

[0091] The description of S401 - S403 can be found in the above embodiment and will not be repeated here.

[0092] S404: Determine the grayscale average value of the plurality of grayscale information.

[0093] The grayscale average value refers to the average value of the grayscale values ​​corresponding to multiple grayscale information.

[0094] In the embodiment of the present disclosure, when the grayscale average value of multiple grayscale information is determined, the grayscale average value can intuitively and concisely characterize the central trend of the grayscale values ​​of the image to be identified, providing a reliable reference basis for the subsequent determination of grayscale description information.

[0095] S405: Determine the grayscale standard deviation of the plurality of grayscale information.

[0096] The grayscale standard deviation refers to a value obtained by performing a standard deviation operation on grayscale values ​​corresponding to multiple grayscale information.

[0097] In the embodiment of the present disclosure, when determining the grayscale standard deviation of multiple grayscale information, the grayscale standard deviation can effectively represent the degree of discreteness of the data set composed of the multiple grayscale information.

[0098] S406: The grayscale average value and the grayscale standard deviation are taken together as grayscale description information.

[0099] In the embodiment of the present disclosure, the grayscale average value can effectively characterize the central trend of the grayscale values ​​of the image to be identified, and the grayscale average values ​​of different images to be identified may be the same. Therefore, when the grayscale average value and the grayscale standard deviation are combined as grayscale description information, the description effect of the grayscale description information can be effectively improved, and the reliability of the obtained grayscale description information can be ensured.

[0100] It is understandable that in the image recognition process, an image recognition sequence of the image to be recognized is usually obtained based on a hash algorithm, and the recognition of the same image to be recognized is achieved based on the image recognition sequence. In this process, the method of increasing the number of bits of the image recognition sequence is adopted to improve the recognition accuracy, which leads to an increase in the computational cost of the recognition process and low computational efficiency for large amounts of images to be recognized.

[0101] Therefore, in the embodiment of the present disclosure, the grayscale mean value and the grayscale standard deviation can be combined on the basis of using a low-bit image recognition sequence to assist in the target recognition process of the image to be recognized.

[0102] like Figure 5 As shown, Figure 5 This is a schematic diagram of a target recognition process proposed in an embodiment of the present disclosure, which includes: converting the image to be recognized into a grayscale image based on an algorithm, performing image segmentation processing on the grayscale image to obtain multiple sub-images, and calculating the grayscale value corresponding to each sub-image; based on the grayscale value corresponding to each sub-image, using a hash algorithm to obtain an image recognition sequence corresponding to the image to be processed, and calculating the grayscale value average and grayscale standard deviation of the multiple sub-images; and then using the obtained image recognition sequence, grayscale value average and grayscale standard deviation as reference data in the target recognition process; when the image recognition sequences, grayscale value averages and grayscale standard deviations corresponding to two images to be recognized are the same, the two images to be recognized are determined to be the same image.

[0103] That is to say, in the embodiment of the present disclosure, after determining multiple grayscale information corresponding to multiple sub-images respectively, the grayscale average value of the multiple grayscale information can be determined, and the grayscale standard deviation of the multiple grayscale information can be determined, and then the grayscale average value and the grayscale standard deviation are used together as grayscale description information. Thus, the obtained grayscale average value can effectively characterize the grayscale value concentration trend of the multiple grayscale information, and the obtained grayscale standard deviation can effectively characterize the degree of grayscale value dispersion of the multiple grayscale information. When the grayscale average value and the grayscale standard deviation are used together as grayscale description information, the grayscale description effect of the grayscale description information on the image to be identified can be effectively improved.

[0104] S407: Obtain a preset grayscale threshold.

[0105] S408: Compare the plurality of grayscale information with the preset grayscale thresholds respectively to obtain a plurality of comparison results.

[0106] The description of S407 and S408 can be found in the above embodiment and will not be repeated here.

[0107] S409: If the comparison result is: the grayscale information is greater than the preset grayscale threshold, it is determined that the value corresponding to the corresponding sub-image identifier is the first recognition value.

[0108] The identification value refers to the value corresponding to the sub-image identifier, determined based on the comparison results. For example, the identification value can be a numeric value such as 0 or 1, or can be represented by any other possible character, such as a or b, without limitation. The first identification value can refer to the value corresponding to the sub-image identifier when the grayscale information is greater than a preset grayscale threshold.

[0109] In the embodiment of the present disclosure, by configuring the value corresponding to the corresponding sub-image identifier as the first identification value when the grayscale information is greater than the preset grayscale threshold, a unified value for the sub-image identifier corresponding to the grayscale information greater than the preset grayscale threshold can be achieved, thereby simplifying the representation form of the sub-image identifier.

[0110] S410: If the comparison result is: the grayscale information is less than or equal to the preset grayscale threshold, determining that the value corresponding to the corresponding sub-image identifier is a second identification value.

[0111] The second identification value refers to the value corresponding to the sub-image identifier when the grayscale information is less than or equal to the preset grayscale threshold.

[0112] In the embodiment of the present disclosure, the second identification value can be distinguished from the first identification value, so as to distinguish sub-image identifiers with different characteristics and provide a reliable reference basis for the subsequent formation of image identification identifiers.

[0113] S411: forming an image recognition sequence according to a plurality of sub-image identifiers and corresponding first recognition values ​​or second recognition values.

[0114] In some embodiments, when forming an image recognition sequence based on multiple sub-image identifiers and corresponding first recognition values ​​or second recognition values, the multiple sub-image identifiers and corresponding first recognition values ​​or second recognition values ​​can be input into a pre-trained machine learning model to obtain the image recognition sequence.

[0115] In other embodiments, when multiple sub-image identifiers and corresponding first identification values ​​or second identification values ​​form an image identification sequence, it is also possible to predetermine the conversion relationship between the multiple sub-image identifiers, the multiple sub-image identifiers' corresponding first identification values ​​or second identification values, and the image identification sequence, and then based on the conversion relationship, form an image identification sequence according to the multiple sub-image identifiers and the corresponding first identification values ​​or second identification values.

[0116] Of course, in some embodiments, when forming an image recognition sequence based on multiple sub-image identifiers and corresponding first recognition values ​​or second recognition values, any other possible method may be used, and there is no limitation to this.

[0117] Optionally, in some embodiments, when forming an image recognition sequence based on multiple sub-image identifiers and corresponding first recognition values ​​or second recognition values, multiple sorting positions corresponding to the multiple sub-image identifiers can be determined, and then the corresponding first recognition values ​​or second recognition values ​​are arranged based on the multiple sorting positions to obtain the image recognition sequence. In this way, the image recognition sequence can neatly and clearly represent the recognition values ​​corresponding to each sub-image, effectively avoiding sorting confusion that affects recognition efficiency and recognition effect, so as to achieve rapid analysis and processing of the image recognition sequence.

[0118] The sorting position refers to the sorting information pre-configured for multiple sub-image identifiers.

[0119] That is to say, after the embodiment of the present disclosure compares multiple grayscale information with the preset grayscale threshold respectively to obtain multiple comparison results, if the comparison result is: the grayscale information is greater than the preset grayscale threshold, then the value corresponding to the corresponding sub-image identifier is determined to be the first recognition value; if the comparison result is: the grayscale information is less than or equal to the preset grayscale threshold, then the value corresponding to the corresponding sub-image identifier is determined to be the second recognition value, and then an image recognition sequence is formed according to the multiple sub-image identifiers and the corresponding first recognition value or the second recognition value. Thus, the grayscale information of the multiple sub-images can be identified and classified based on the preset grayscale threshold, so that the obtained image recognition sequence can concisely and clearly characterize the grayscale features of each sub-image, thereby effectively reducing the amount of data of the image recognition sequence during the image recognition process, and can effectively improve the image recognition efficiency while ensuring the image recognition effect.

[0120] For example, the preset grayscale threshold can be the grayscale average of multiple grayscale information, and then the grayscale information corresponding to each sub-image is compared with the preset grayscale threshold. When the value of the grayscale information is greater than the preset grayscale value, the comparison result is taken as 1. When the value of the grayscale information is less than or equal to the preset grayscale value, the comparison result is taken as 0. Then, the multiple comparison results are spliced ​​according to the order of the sub-images to obtain a character string containing 0 and 1, and it is used as the image recognition sequence, for example, it can be 1000010...0100, and the number of bits of the character string can be equal to the number of sub-images.

[0121] S412: Perform target recognition on the image to be recognized based on the grayscale description information and the image recognition sequence.

[0122] The description of S412 can be found in the above embodiment and will not be repeated here.

[0123] In this embodiment, by determining the grayscale average value of multiple grayscale information and determining the grayscale standard deviation of multiple grayscale information, and then using the grayscale average value and the grayscale standard deviation together as grayscale description information, the obtained grayscale average value can effectively characterize the grayscale value concentration trend of multiple grayscale information, and the obtained grayscale standard deviation can effectively characterize the degree of grayscale value dispersion of multiple grayscale information. When the grayscale average value and the grayscale standard deviation are used together as grayscale description information, the grayscale description effect of the grayscale description information on the image to be identified can be effectively improved. When the comparison result indicates that the grayscale information is greater than a preset grayscale threshold, the value corresponding to the corresponding sub-image identifier is determined to be the first identification value. When the comparison result indicates that the grayscale information is less than or equal to the preset grayscale threshold, the value corresponding to the corresponding sub-image identifier is determined to be the second identification value. An image recognition sequence is then formed based on the multiple sub-image identifiers and the corresponding first identification values ​​or second identification values. Thus, the grayscale information of the multiple sub-images can be identified and classified based on the preset grayscale threshold, so that the resulting image recognition sequence can concisely and clearly represent the grayscale features of each sub-image, thereby effectively reducing the amount of data in the image recognition sequence during the image recognition process, effectively improving image recognition efficiency while ensuring image recognition results. By determining multiple sorting positions corresponding to the multiple sub-image identifiers, and then arranging the corresponding first identification values ​​or second identification values ​​based on the multiple sorting positions, an image recognition sequence is obtained. Thus, the image recognition sequence neatly and clearly represents the identification values ​​corresponding to each sub-image, effectively avoiding sorting confusion that affects recognition efficiency and results, thereby achieving rapid analysis and processing of the image recognition sequence.

[0124] Figure 6 It is a flowchart of an image recognition method proposed in another embodiment of the present disclosure.

[0125] like Figure 6 As shown, the image recognition method includes:

[0126] S601: Acquire an image to be recognized, wherein there are multiple images to be recognized.

[0127] S602: Process the image to be identified to obtain grayscale description information.

[0128] S603: Determine an image recognition sequence according to the grayscale description information.

[0129] The description of S601-S603 can be found in the above embodiment and will not be repeated here.

[0130] S604: storing the image to be recognized according to the grayscale description information and the image recognition sequence.

[0131] It is understandable that the image to be identified is usually unstructured data, which results in a large amount of storage space occupied by the image to be identified and a relatively cumbersome analysis and processing process. Grayscale description information and image recognition sequences are structured data and can effectively characterize the feature information of the image to be identified and meet the information requirements of the subsequent target recognition process.

[0132] That is to say, after obtaining the grayscale description information and the image recognition sequence, the embodiment of the present disclosure can store the image to be identified based on the grayscale description information and the image recognition sequence. Since the image to be identified may be unstructured data, when the image to be identified is stored based on the grayscale description information and the image recognition sequence corresponding to the image to be identified, the structured conversion of the image data to be identified can be achieved while ensuring the representation ability of the image to be identified, which can effectively reduce the storage space occupied by the image to be identified and assist in the creation of the image relationship network.

[0133] S605: If the first image to be recognized and the second image to be recognized meet a set condition, construct an image relationship network based on the first image to be recognized and the second image to be recognized, wherein the image relationship network provides an image retrieval service.

[0134] The set condition refers to a condition pre-configured based on the relevant information between the first image to be identified and the second image to be identified, which can be used to determine whether to construct an image relationship network based on the first image to be identified and the second image to be identified.

[0135] The image relationship network refers to a network constructed based on multiple images to be identified that meet preset conditions. Multiple images to be identified in the same relationship network can meet the preset conditions at the same time.

[0136] In the embodiment of the present disclosure, the preset conditions may include: the similarity between the first image to be identified and the second image to be identified is greater than 0.8, and the similarity can be determined based on the grayscale average value, grayscale standard deviation and image recognition sequence of the first image to be identified and the second image to be identified, or the preset conditions may also include: the grayscale variance of the first image to be identified and the second image to be identified is the same, and the grayscale extreme values ​​of the first image to be identified and the second image to be identified are the same, and there is no limitation on this.

[0137] Optionally, in some embodiments, the above-mentioned setting conditions include: the grayscale average value of the first image to be identified and the grayscale average value of the second image to be identified are the same, the grayscale standard deviation of the first image to be identified and the grayscale standard deviation of the second image to be identified are the same, and the image recognition sequence of the first image to be identified and the image recognition sequence of the second image to be identified are the same. Since the grayscale features of the image have high randomness, when the grayscale average values, grayscale standard deviations, and image recognition sequences corresponding to the first image to be identified and the second image to be identified are the same, it is characterized that the first image to be identified and the second image to be identified are more likely to be the same image. Therefore, the setting conditions can provide a reliable trigger basis for constructing an image relationship network, thereby ensuring the accuracy of the obtained image relationship network.

[0138] In the process of constructing the above-mentioned image relationship network, by comparing the grayscale description information and image recognition sequences of different images to be identified, registrants using the same qualification images can be detected, and association relationships can be established to facilitate statistical analysis. For example, Figure 7 As shown, Figure 7 This is a schematic diagram of the target recognition process for the image to be identified in the embodiment of the present disclosure. The qualification images of registrants ①, ②, and ③ are processed separately to obtain corresponding grayscale description information and image recognition sequences. The grayscale description information and image recognition sequences corresponding to registrants ①, ②, and ③ are then analyzed and compared to determine whether a corresponding image relationship network has been established. This image recognition method can be used for target recognition of qualification images to identify and control risky images.

[0139] That is to say, in the embodiment of the present disclosure, the number of images to be identified can be multiple. After determining the image recognition sequence based on the grayscale description information, if the set conditions are met between the first image to be identified and the second image to be identified, an image relationship network is constructed based on the first image to be identified and the second image to be identified, wherein the image relationship network provides image retrieval services. When the set conditions are met between the first image to be identified and the second image to be identified, there may be a high correlation between the first image to be identified and the second image to be identified. At this time, by constructing an image relationship network based on the first image to be identified and the second image to be identified, the creation of an unstructured image data relationship network can be realized. The obtained image relationship network can provide image retrieval services to quickly and accurately identify multiple images with high correlation from a large amount of image data, which can effectively make up for risk loopholes.

[0140] In this embodiment, by storing the images to be identified based on grayscale description information and image recognition sequences, since the images to be identified may be unstructured data, storing the images to be identified based on their corresponding grayscale description information and image recognition sequences can achieve structured conversion of the image data to be identified while maintaining the image representation capability, effectively reducing the storage space occupied by the images to be identified and assisting in the creation of an image relationship network. There may be multiple images to be identified. After determining the image recognition sequence based on the grayscale description information, if a set condition is met between the first image to be identified and the second image to be identified, an image relationship network is constructed based on the first image to be identified and the second image to be identified. The image relationship network provides image retrieval services. When the set condition is met between the first image to be identified and the second image to be identified, there may be a high correlation between the first image to be identified and the second image to be identified. In this case, constructing an image relationship network based on the first image to be identified and the second image to be identified can achieve the creation of an unstructured image data relationship network. The resulting image relationship network can provide image retrieval services to quickly and accurately identify multiple images with high correlation from a large amount of image data, effectively addressing risk vulnerabilities. The setting conditions may include: the grayscale average value of the first image to be identified is the same as the grayscale average value of the second image to be identified, the grayscale standard deviation of the first image to be identified is the same as the grayscale standard deviation of the second image to be identified, and the image recognition sequence of the first image to be identified is the same as the image recognition sequence of the second image to be identified. Since the grayscale features of the image have a high degree of randomness, when the grayscale average values, grayscale standard deviations, and image recognition sequences corresponding to the first image to be identified and the second image to be identified are the same, it indicates that the first image to be identified and the second image to be identified are likely to be the same image. Therefore, the setting conditions can provide a reliable trigger basis for constructing an image relationship network, thereby ensuring the accuracy of the obtained image relationship network.

[0141] Figure 8Schematic diagram of the structure of an image recognition device proposed in one embodiment of the present disclosure.

[0142] like Figure 8 As shown, the image recognition device 80 includes:

[0143] An acquisition module 801 is used to acquire an image to be recognized;

[0144] The processing module 802 is used to process the image to be identified and obtain grayscale description information;

[0145] A determination module 803 is used to determine an image recognition sequence based on the grayscale description information;

[0146] The recognition module 804 is used to perform target recognition on the image to be recognized based on the grayscale description information and the image recognition sequence.

[0147] In some embodiments of the present disclosure, Figure 9 As shown, Figure 9 FIG. 8 is a schematic diagram of the structure of an image recognition device proposed in another embodiment of the present disclosure, wherein the processing module 802 includes:

[0148] The first processing submodule 8021 is used to segment the image to be recognized to obtain multiple sub-images;

[0149] A determination submodule 8022 is configured to determine a plurality of grayscale information corresponding to the plurality of sub-images;

[0150] The first generating submodule 8023 is configured to generate grayscale description information according to multiple grayscale information.

[0151] In some embodiments of the present disclosure, the first generating submodule 8023 is specifically configured to:

[0152] Determine the grayscale average value of multiple grayscale information;

[0153] Determine the grayscale standard deviation of multiple grayscale information;

[0154] The grayscale mean and grayscale standard deviation are used together as grayscale description information.

[0155] In some embodiments of the present disclosure, the determination module 803 includes:

[0156] An acquisition submodule 8031 ​​is used to obtain a preset grayscale threshold;

[0157] The second processing submodule 8032 is used to compare the multiple grayscale information with the preset grayscale thresholds respectively to obtain multiple comparison results;

[0158] The second generating submodule 8033 is configured to form an image recognition sequence according to the multiple comparison results and the multiple sub-image identifiers corresponding thereto, wherein the sub-image identifiers are used to identify corresponding sub-images.

[0159] In some embodiments of the present disclosure, the second generation submodule 8033 is specifically configured to:

[0160] When the comparison result is: the grayscale information is greater than the preset grayscale threshold, determining that the value corresponding to the corresponding sub-image identifier is the first recognition value;

[0161] When the comparison result is that the grayscale information is less than or equal to the preset grayscale threshold, determining that the value corresponding to the corresponding sub-image identifier is the second identification value;

[0162] An image recognition sequence is formed according to the multiple sub-image identifiers and the corresponding first recognition values ​​or second recognition values.

[0163] In some embodiments of the present disclosure, the second generation submodule 8033 is further configured to:

[0164] Determining a plurality of sorting positions corresponding to the plurality of sub-image identifiers;

[0165] The corresponding first identification values ​​or second identification values ​​are arranged based on the multiple sorting positions to obtain an image identification sequence.

[0166] In some embodiments of the present disclosure, the number of images to be identified is multiple;

[0167] The identification module 804 is specifically configured to:

[0168] When a set condition is satisfied between the first image to be recognized and the second image to be recognized, an image relationship network is constructed based on the first image to be recognized and the second image to be recognized, wherein the image relationship network provides an image retrieval service.

[0169] In some embodiments of the present disclosure, the setting conditions include:

[0170] The grayscale average value of the first image to be identified is the same as the grayscale average value of the second image to be identified;

[0171] The grayscale standard deviation of the first image to be identified is the same as the grayscale standard deviation of the second image to be identified;

[0172] The image recognition sequence of the first image to be recognized is the same as the image recognition sequence of the second image to be recognized.

[0173] In some embodiments of the present disclosure, the device further includes:

[0174] The storage module 805 is used to store the image to be recognized according to the grayscale description information and the image recognition sequence.

[0175] It should be noted that the aforementioned explanation of the image recognition method is also applicable to the image recognition device of this embodiment and will not be repeated here.

[0176] In this embodiment, by acquiring the image to be identified and processing the image to be identified, grayscale description information is obtained, and based on the grayscale description information, an image recognition sequence is determined, and based on the grayscale description information and the image recognition sequence, target recognition is performed on the image to be identified. Thus, by acquiring the grayscale description information and the image recognition sequence of the image to be identified to perform image recognition, a reliable reference basis can be provided for the target recognition process of the image to be identified, thereby effectively improving the target recognition effect of the image to be identified.

[0177] Figure 10 A block diagram of an exemplary computer device suitable for implementing embodiments of the present disclosure is shown. Figure 10 The computer device 12 shown is only an example and should not bring any limitation to the functionality and scope of use of the embodiments of the present disclosure.

[0178] like Figure 10 As shown, computer device 12 is implemented as a general-purpose computing device. Components of computer device 12 may include, but are not limited to, one or more processors or processing units 16, system memory 28, and a bus 18 that connects various system components (including system memory 28 and processing unit 16).

[0179] Bus 18 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor, or a local bus using any of a variety of bus architectures. Examples of such architectures include, but are not limited to, the Industry Standard Architecture (ISA) bus, the Micro Channel Architecture (MAC) bus, the Enhanced ISA bus, the Video Electronics Standards Association (VESA) local bus, and the Peripheral Component Interconnection (PCI) bus.

[0180] The computer device 12 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by the computer device 12, including volatile and non-volatile media, removable and non-removable media.

[0181] The memory 28 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 30 and / or cache memory 32. The computer device 12 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, the storage system 34 may be configured to read and write non-removable, non-volatile magnetic media ( Figure 10 Not shown, often called a "hard drive").

[0182] although Figure 10 Although not shown, a disk drive for reading and writing to a removable non-volatile disk (e.g., a "floppy disk"), and an optical disk drive for reading and writing to a removable non-volatile optical disk (e.g., a Compact Disc Read Only Memory (hereinafter referred to as: CD-ROM), a Digital Video Disc Read Only Memory (hereinafter referred to as: DVD-ROM), or other optical media) may be provided. In these cases, each drive may be connected to the bus 18 via one or more data medium interfaces. The memory 28 may include at least one program product having a set (e.g., at least one) of program modules configured to perform the functions of the various embodiments of the present disclosure.

[0183] A program / utility 40 having a set (at least one) of program modules 42 may be stored, for example, in memory 28. Such program modules 42 include, but are not limited to, an operating system, one or more application programs, other program modules, and program data, each of which, or some combination thereof, may include an implementation of a network environment. Program modules 42 generally implement the functions and / or methods of the embodiments described herein.

[0184] The computer device 12 can also communicate with one or more external devices 14 (e.g., a keyboard, pointing device, display 24, etc.), one or more devices that enable human interaction with the computer device 12, and / or any device that enables the computer device 12 to communicate with one or more other computing devices (e.g., a network card, a modem, etc.). This communication can occur via an input / output (I / O) interface 22. Furthermore, the computer device 12 can communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network such as the Internet) via a network adapter 20. As shown, the network adapter 20 communicates with the other modules of the computer device 12 via a bus 18. It should be understood that, although not shown, other hardware and / or software modules can be used in conjunction with the computer device 12, including but not limited to microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0185] The processing unit 16 executes various functional applications and image recognition by running programs stored in the system memory 28 , such as implementing the image recognition method mentioned in the above embodiments.

[0186] In order to implement the above embodiments, the present disclosure further proposes a non-transitory computer-readable storage medium having a computer program stored thereon. When the program is executed by a processor, the image recognition method proposed in the above embodiments of the present disclosure is implemented.

[0187] In order to implement the above embodiments, the present disclosure further proposes a computer program product. When a processor executes instructions in the computer program product, the image recognition method proposed in the above embodiments of the present disclosure is executed.

[0188] Other embodiments of the present disclosure will readily occur to those skilled in the art after considering the specification and practicing the invention disclosed herein. This disclosure is intended to cover any variations, uses, or adaptations of the present disclosure that follow from the general principles of the present disclosure and include common knowledge or customary techniques in the art not disclosed herein. The description and examples are to be considered as exemplary only, with the true scope and spirit of the present disclosure being indicated by the following claims.

[0189] It should be understood that the present disclosure is not limited to the exact structures that have been described above and shown in the drawings, and that various modifications and changes can be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

[0190] It should be noted that, in the description of this disclosure, the terms "first", "second", etc. are used for descriptive purposes only and should not be understood as indicating or implying relative importance. In addition, in the description of this disclosure, unless otherwise specified, the meaning of "plurality" is two or more.

[0191] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code that includes one or more executable instructions for implementing the steps of a specific logical function or process, and the scope of the preferred embodiments of the present disclosure includes additional implementations in which functions may be performed out of the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by those skilled in the art to which the embodiments of the present disclosure belong.

[0192] It should be understood that various parts of the present disclosure can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented using hardware, as in another embodiment, any one of the following technologies known in the art or a combination thereof can be used to implement: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application-specific integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0193] Those skilled in the art will understand that all or part of the steps in the method of the above embodiment can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium. When the program is executed, it includes one or a combination of the steps of the method embodiment.

[0194] In addition, the functional units in the various embodiments of the present disclosure may be integrated into a single processing module, or each unit may exist physically separately, or two or more units may be integrated into a single module. The aforementioned integrated modules may be implemented in the form of hardware or in the form of software functional modules. If the integrated modules are implemented in the form of software functional modules and sold or used as independent products, they may also be stored in a computer-readable storage medium.

[0195] The storage medium mentioned above can be a read-only memory, a magnetic disk or an optical disk, etc.

[0196] Throughout this specification, reference to terms such as "one embodiment," "some embodiments," "examples," "specific examples," or "some examples" means that the specific features, structures, materials, or characteristics described in conjunction with that embodiment or example are included in at least one embodiment or example of the present disclosure. In this specification, schematic representations of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

[0197] Although the embodiments of the present disclosure have been shown and described above, it is understood that the above embodiments are illustrative and are not to be construed as limitations on the present disclosure. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present disclosure.

Claims

1. An image recognition method, characterized in that: include: Acquire an image to be identified, where the image to be identified is unstructured data; Performing image segmentation on the image to be recognized to obtain a plurality of sub-images, and taking the grayscale average value and grayscale standard deviation of the plurality of grayscale information as grayscale description information according to the plurality of grayscale information corresponding to the plurality of sub-images; Comparing the plurality of grayscale information with a preset grayscale threshold value respectively, and determining that the value corresponding to the corresponding sub-image identifier is a first identification value if the grayscale information is greater than the preset grayscale threshold value; and determining that the value corresponding to the corresponding sub-image identifier is a second identification value if the grayscale information is less than or equal to the preset grayscale threshold value; forming an image identification sequence according to the multiple sub-image identifiers and the corresponding first identification value or the second identification value, wherein the sub-image identifier is used to identify the corresponding sub-image; storing the image to be identified according to the grayscale description information and the image recognition sequence; Target recognition is performed on the images to be recognized based on the grayscale description information and the image recognition sequence; wherein, if the grayscale mean value, grayscale standard deviation and image recognition sequence of the first image to be recognized and the second image to be recognized in the images to be recognized are the same, an image relationship network for providing image retrieval services is constructed based on the first image to be recognized and the second image to be recognized.

2. The method according to claim 1, wherein The forming of the image recognition sequence according to the plurality of sub-image identifiers and the corresponding first recognition value or the second recognition value includes: Determining a plurality of sorting positions corresponding to the plurality of sub-image identifiers respectively; The corresponding first identification values ​​or the second identification values ​​are arranged based on the multiple sorting positions to obtain the image identification sequence.

3. An image recognition device, characterized in that: include: An acquisition module is used to acquire an image to be identified, where the image to be identified is unstructured data; a processing module configured to segment the image to be identified to obtain a plurality of sub-images, and to use a grayscale average value and a grayscale standard deviation of the plurality of grayscale information as grayscale description information based on a plurality of grayscale information corresponding to the plurality of sub-images; a determination module, configured to compare the plurality of grayscale information with a preset grayscale threshold value to obtain a plurality of comparison results, and determine that the value corresponding to the corresponding sub-image identifier is a first identification value if the grayscale information is greater than the preset grayscale threshold value; and determine that the value corresponding to the corresponding sub-image identifier is a second identification value if the grayscale information is less than or equal to the preset grayscale threshold value; forming an image identification sequence according to the multiple sub-image identifiers and the corresponding first identification value or the second identification value, wherein the sub-image identifier is used to identify the corresponding sub-image; a storage module, configured to store the image to be identified according to the grayscale description information and the image recognition sequence; The recognition module is used to perform target recognition on the image to be recognized based on the grayscale description information and the image recognition sequence; wherein, if the grayscale mean value, grayscale standard deviation and image recognition sequence of the first image to be recognized and the second image to be recognized are the same, an image relationship network for providing image retrieval services is constructed based on the first image to be recognized and the second image to be recognized.

4. The device according to claim 3, characterized in that Forming the image recognition sequence according to the multiple sub-image identifiers and the corresponding first recognition values ​​or the second recognition values ​​includes: Determining a plurality of sorting positions corresponding to the plurality of sub-image identifiers respectively; The corresponding first identification values ​​or the second identification values ​​are arranged based on the multiple sorting positions to obtain the image identification sequence.

5. A computer device, characterized in that: include: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method of claim 1 or 2.

6. A non-transitory computer-readable storage medium storing computer instructions, characterized in that: in, The computer instructions are used to cause the computer to execute the method according to claim 1 or 2.

7. A computer program product, characterized in that The invention comprises a computer program which, when executed by a processor, implements the steps of the method according to claim 1 or 2.

Images