A method, system and storage medium for identifying and processing user images in a private domain e-commerce platform

Abstract

The present application provides a method, system and storage medium for identifying and processing user images in a private domain e-commerce platform. The image resources generated by the target user on the platform are decomposed and preprocessed to obtain thumbnail images, and the hash value of the thumbnail image is calculated to generate hash value data to be tested. The hash value data to be tested is converted into a numerical value of an interval to be tested according to the image information, and the interval value to be tested is used to search for related comparison image resources in the cloud database, and the image similarity of the target user is judged by comparing the hash values ​​of the image resources to be tested and the comparison image resources. The present application effectively improves the accuracy by optimizing and adjusting various indicators of the open source image recognition algorithm; at the same time, an optimization step is added during the comparison process, and the hash value calculated for the image is converted into an interval value again for filtering to improve the image retrieval performance. Recognition and comparison can be performed in parallel according to different interval values, which significantly improves the comparison performance.

Description

Technical Field

[0001] The present application belongs to the field of computer analysis technology, and more specifically, relates to a method, system and storage medium for identifying and processing user images in a private domain e-commerce platform. Background Art

[0002] With the development of the Internet, e-commerce has brought great changes to our lives. People are becoming more and more dependent on e-commerce and are beginning to influence e-commerce through social channels. Private domain e-commerce has emerged as a new model that combines social and e-commerce. However, the combination of social and e-commerce has created a series of problems under this model. For example, as the amount of data increases, the performance of existing processing technologies decreases. When it reaches a certain level, the scenario of pursuing fast matching in the Internet industry is almost unavailable. Summary of the invention

[0003] In view of this, the present application provides a method, system and storage medium for identifying and processing user images in a private domain e-commerce platform, which supports horizontal expansion, and the identification and matching speed increases with the expansion, significantly improving the data identification and processing efficiency.

[0004] The specific technical solutions of this application are as follows:

[0005] The first aspect of the present application provides a method for identifying and processing user images in a private domain e-commerce platform, comprising the following steps:

[0006] Collect all the image resources generated by the target user on platform A, and decompose and pre-process the collected image resources to obtain thumbnail images;

[0007] Calculate the hash value of the thumbnail image to generate batch hash value data, convert the batch hash value data into batch interval values ​​according to the image information, and store the batch hash value data and the batch interval values ​​in the cloud database;

[0008] Obtain the image resources to be tested generated by the target user on platform B, calculate the hash value data to be tested of the image and generate the corresponding interval value to be tested;

[0009] The numerical value of the interval to be tested is used to search for the associated comparison picture resources in the cloud database, and the similarity of the target user's pictures is determined by comparing the hash values ​​of the picture resources to be tested and the comparison picture resources.

[0010] Preferably, the collected image resources are decomposed and preprocessed as follows:

[0011] Calculate the RGB color component values ​​of the image resource and determine whether it exceeds the threshold;

[0012] When the RGB color component value exceeds the threshold, the data size of the image resource is adjusted to uniformly limit the RGB color components of the image resource to within the threshold.

[0013] Preferably, it also includes:

[0014] Obtain the log information generated by the target user under the platform operation node, and filter the data packets with graphic tags in the log information, where the operation nodes include upload, download and forward;

[0015] The data resources are cleaned according to the source information, process information and security information in the data packet to obtain the image resources of the target user in the platform.

[0016] Preferably, the hash value of the thumbnail image is calculated to generate batch hash value data as follows:

[0017] Get the RGB value of each pixel in the thumbnail image, where the information of the second highest (16-23) red component is taken out to get the r value, the information of the middle (8-15) green component is taken out to get the g value, and the information of the low (0-7) blue component is taken out to get the b value;

[0018] The grayscale value of the image is calculated according to the RGB value of the image, and the RGB value and grayscale value of the image are used to generate the hash value data of the image using the difference hash algorithm.

[0019] Preferably, searching the cloud database for the associated comparison picture resources using the value of the interval to be tested is specifically as follows:

[0020] Obtain static information of target users, analyze preference information and application information of target users, and set different weight ratios for values ​​in different intervals according to preference information and application information;

[0021] Retrieve interval data and compare image resources whose relevance is still within the interval range after introducing different weight calculations.

[0022] Preferably, the batch hash value data is converted into batch interval values ​​according to the image information as follows:

[0023] Partition the hash value data according to the attribute information and function information in the image information, and convert the hash value data into a numerical value in the range of 0 to 32;

[0024] Multiple intervals are created for interval values ​​according to a preset allocation method, and the intervals are allocated to different system nodes.

[0025] Preferably, storing batch hash value data and batch interval values ​​into the cloud database is specifically as follows:

[0026] Send batch interval values ​​to the db and introduce index labels to them;

[0027] Store the batch hash value data in redis and establish a mapping relationship with the batch interval value.

[0028] Preferably, storing batch hash value data and batch interval values ​​into the cloud database is specifically as follows:

[0029] Collect digital information of batch interval values, format the digital information and generate key data;

[0030] Use key data to encrypt redis when storing batch hash value data and batch interval value data.

[0031] Preferably, the similarity of the target user's pictures is determined by comparing the hash values ​​of the picture resource to be tested and the comparison picture resource as follows:

[0032] Obtain hash value data of the image resource to be tested and the image resource to be compared respectively, analyze the data elements and data structure in the hash value data, and arrange the data elements and data structure in order;

[0033] The data distance between the image resource to be tested and the image resource to be compared is counted according to the arrangement, and the image similarity of the target user is determined based on the threshold of the data distance.

[0034] In a second aspect, the present application provides a recognition processing system based on user images in a private domain e-commerce platform, comprising a memory and a processor, wherein the memory comprises a recognition processing program based on user images in the private domain e-commerce platform, and when the recognition processing program based on user images in the private domain e-commerce platform is executed by the processor, the following steps are implemented:

[0035] Collect all the image resources generated by the target user on platform A, and decompose and pre-process the collected image resources to obtain thumbnail images;

[0036] Calculate the hash value of the thumbnail image to generate batch hash value data, convert the batch hash value data into batch interval values ​​according to the image information, and store the batch hash value data and the batch interval values ​​in the cloud database;

[0037] Obtain the image resources to be tested generated by the target user on platform B, calculate the hash value data to be tested of the image and generate the corresponding interval value to be tested;

[0038] The numerical value of the interval to be tested is used to search for the associated comparison picture resources in the cloud database, and the similarity of the target user's pictures is determined by comparing the hash values ​​of the picture resources to be tested and the comparison picture resources.

[0039] Preferably, the collected image resources are decomposed and preprocessed as follows:

[0040] Calculate the RGB color component values ​​of the image resource and determine whether it exceeds the threshold;

[0041] When the RGB color component value exceeds the threshold, the data size of the image resource is adjusted to uniformly limit the RGB color components of the image resource to within the threshold.

[0042] Preferably, it also includes:

[0043] Obtain the log information generated by the target user under the platform operation node, and filter the data packets with graphic tags in the log information, where the operation nodes include upload, download and forward;

[0044] The data resources are cleaned according to the source information, process information and security information in the data packet to obtain the image resources of the target user in the platform.

[0045] Preferably, the hash value of the thumbnail image is calculated to generate batch hash value data as follows:

[0046] Get the RGB value of each pixel in the thumbnail image, where the information of the second highest (16-23) red component is taken out to get the r value, the information of the middle (8-15) green component is taken out to get the g value, and the information of the low (0-7) blue component is taken out to get the b value;

[0047] The grayscale value of the image is calculated according to the RGB value of the image, and the RGB value and grayscale value of the image are used to generate the hash value data of the image using the difference hash algorithm.

[0048] Preferably, searching the cloud database for the associated comparison picture resources using the value of the interval to be tested is specifically as follows:

[0049] Obtain static information of target users, analyze preference information and application information of target users, and set different weight ratios for values ​​in different intervals according to preference information and application information;

[0050] Retrieve interval data and compare image resources whose relevance is still within the interval range after introducing different weight calculations.

[0051] Preferably, the batch hash value data is converted into batch interval values ​​according to the image information as follows:

[0052] Partition the hash value data according to the attribute information and function information in the image information, and convert the hash value data into a numerical value in the range of 0 to 32;

[0053] Multiple intervals are created for interval values ​​according to a preset allocation method, and the intervals are allocated to different system nodes.

[0054] Preferably, storing batch hash value data and batch interval values ​​into the cloud database is specifically as follows:

[0055] Send batch interval values ​​to the db and introduce index labels to them;

[0056] Store the batch hash value data in redis and establish a mapping relationship with the batch interval value.

[0057] Preferably, storing batch hash value data and batch interval values ​​into the cloud database is specifically as follows:

[0058] Collect digital information of batch interval values, format the digital information and generate key data;

[0059] Use key data to encrypt redis when storing batch hash value data and batch interval value data.

[0060] Preferably, the similarity of the target user's pictures is determined by comparing the hash values ​​of the picture resource to be tested and the comparison picture resource as follows:

[0061] Obtain hash value data of the image resource to be tested and the image resource to be compared respectively, analyze the data elements and data structure in the hash value data, and arrange the data elements and data structure in order;

[0062] The data distance between the image resource to be tested and the image resource to be compared is counted according to the arrangement, and the image similarity of the target user is determined based on the threshold of the data distance.

[0063] The third aspect of the present application provides a computer-readable storage medium, which includes an identification processing program based on user pictures in a private domain e-commerce platform. When the identification processing program based on user pictures in the private domain e-commerce platform is executed by a processor, the steps of the identification processing method based on user pictures in the private domain e-commerce platform are implemented.

[0064] In summary, the present application provides a method, system and storage medium for identifying and processing user images in a private domain e-commerce platform. The image resources generated by the target user on the platform are decomposed and preprocessed to obtain thumbnail images, and the hash value of the thumbnail image is calculated to generate hash value data to be tested. The hash value data to be tested is converted into a numerical value of an interval to be tested according to the image information, and the interval value to be tested is used to search for related comparison image resources in the cloud database, and the image similarity of the target user is judged by comparing the hash values ​​of the image resources to be tested and the comparison image resources. The present application effectively improves the accuracy by optimizing and adjusting various indicators of the open source image recognition algorithm; at the same time, an optimization step is added during the comparison process, and the hash value calculated for the image is converted into an interval value again for filtering to improve the image retrieval performance. Recognition and comparison can be performed in parallel according to different interval values, which significantly improves the comparison performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0066] Figure 1 This is a flowchart of a method for identifying and processing user images in a private domain e-commerce platform for this application;

[0067] Figure 2 This application provides a block diagram of a system for identifying and processing user images in a private domain e-commerce platform. DETAILED DESCRIPTION

[0068] In order to make the purpose, features, and advantages of this application more obvious and easy to understand, the technical solutions in the embodiments of this application are clearly and completely described. Obviously, the embodiments described below are only part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0069] Please refer to Figure 1 , Figure 1 This application provides a flowchart of a method for identifying and processing user images in a private domain e-commerce platform.

[0070] The first aspect of the embodiment of the present application provides a method for identifying and processing user pictures in a private domain e-commerce platform, characterized in that it includes the following steps:

[0071] S102: Collect all picture resources generated by the target user on platform A, and decompose and pre-process the collected picture resources to obtain thumbnail images;

[0072] S104: Calculate the hash value of the thumbnail image to generate batch hash value data, convert the batch hash value data into batch interval values ​​according to the image information, and store the batch hash value data and the batch interval values ​​in the cloud database;

[0073] S106: Obtain the image resource to be tested generated by the target user on platform B, calculate the hash value data to be tested of the image and generate the corresponding interval value to be tested;

[0074] S108: Searching for related comparison picture resources in the cloud database using the value of the interval to be tested, and determining the picture similarity of the target user by comparing the hash values ​​of the picture resource to be tested and the comparison picture resource.

[0075] It should be noted that, assuming that the user data of platform A is isolated from that of platform B, the image resources of the platform can be the operation data generated by the target user under the platform operation node. Using the interval data of the image resource as an index to search for the corresponding image resource, and then retrieving the hash value of the image resource as the value for comparison, can increase the recognition and matching speed and significantly improve the data recognition processing efficiency.

[0076] According to the embodiment of the present application, the collected image resources are decomposed and preprocessed specifically as follows:

[0077] Calculate the RGB color component values ​​of the image resource and determine whether it exceeds the threshold;

[0078] When the RGB color component value exceeds the threshold, the data size of the image resource is adjusted to uniformly limit the RGB color components of the image resource to within the threshold.

[0079] It should be noted that adjusting the image resources to the form of thumbnails facilitates subsequent data analysis and processing. In the embodiment of the present application, the image resources can be reduced to an 8x9 (height x width) image with 8-bit RGB color components.

[0080] According to an embodiment of the present application, it also includes:

[0081] Obtain the log information generated by the target user under the platform operation node, and filter the data packets with graphic tags in the log information, where the operation nodes include upload, download and forward;

[0082] The data resources are cleaned according to the source information, process information and security information in the data packet to obtain the image resources of the target user in the platform.

[0083] It should be noted that by analyzing the user's behavior preferences and potential purposes based on the user's operation information, it is possible to extract image resources that can truly reflect the user's behavior and filter out interference information such as system advertisements.

[0084] According to the embodiment of the present application, the hash value of the thumbnail image is calculated to generate batch hash value data as follows:

[0085] Get the RGB value of each pixel in the thumbnail image, where the information of the second highest (16-23) red component is taken out to get the r value, the information of the middle (8-15) green component is taken out to get the g value, and the information of the low (0-7) blue component is taken out to get the b value;

[0086] The grayscale value of the image is calculated according to the RGB value of the image, and the RGB value and grayscale value of the image are used to generate the hash value data of the image using the difference hash algorithm.

[0087] It should be noted that the formula for calculating the gray value in the embodiment of the present application is: (r*30+g*59+b*11) / 100. The difference hash algorithm is used in the embodiment of the present application to calculate the hash value data, which has the advantages of high efficiency, high accuracy and wide application range. Among them, the difference hash algorithm mainly adopts the principle of difference value calculation, subtracts two adjacent elements in each row of the matrix to obtain several difference values, and processes the difference values. For example, if it is a positive number or 0, it is remembered as 1; if it is a negative number, it is recorded as 0. Then the results are combined together to obtain a hash value.

[0088] According to an embodiment of the present application, searching for the associated comparison picture resources in the cloud database using the value of the interval to be tested is specifically as follows:

[0089] Obtain static information of target users, analyze preference information and application information of target users, and set different weight ratios for values ​​in different intervals according to preference information and application information;

[0090] Retrieve interval data and compare image resources whose relevance is still within the interval range after introducing different weight calculations.

[0091] It should be noted that static information includes user registration information, setting information and order information. By calculating the weights of the interval data of the user's image in different scenarios according to the user's special preference information, the comparison objects can be screened in a targeted manner, the influence of background values ​​can be eliminated, and the analysis results can be closer to the actual situation.

[0092] According to an embodiment of the present application, the batch hash value data is converted into batch interval values ​​according to the image information as follows:

[0093] Partition the hash value data according to the attribute information and function information in the image information, and convert the hash value data into a numerical value in the range of 0 to 32;

[0094] Multiple intervals are created for interval values ​​according to a preset allocation method, and the intervals are allocated to different system nodes.

[0095] It should be noted that the hash value data is partitioned according to the attribute information and function information of the image, and the hash value is converted into interval data. On the one hand, it is convenient to retrieve, analyze and match from multiple aspects and dimensions, and on the other hand, it improves data processing efficiency and reliability. The interval division in the embodiment of the present application is that a node holds only one interval. In the case of n nodes, the value interval of the partition key is evenly divided into n parts, and then each node holds one piece. If the data distribution is uneven, the interval distribution can be adjusted to achieve uniformity, so that data migration will also be reduced.

[0096] According to the embodiment of the present application, the batch hash value data and the batch interval value data are stored in the cloud database specifically as follows:

[0097] Send batch interval values ​​to the db and introduce index labels to them;

[0098] Store the batch hash value data in redis and establish a mapping relationship with the batch interval value.

[0099] It should be noted that the batch interval data reflects the sample values ​​of the label images related to the target user, and the introduction of index labels to improve the retrieval efficiency and accuracy. By establishing a mapping relationship between batch hash values ​​and batch interval values, the retrieval results can be quickly reflected.

[0100] According to the embodiment of the present application, the batch hash value data and the batch interval value data are stored in the cloud database specifically as follows:

[0101] Collect digital information of batch interval values, format the digital information and generate key data;

[0102] Use key data to encrypt redis when storing batch hash value data and batch interval value data.

[0103] It should be noted that using the generated batch interval value as the password stored in redis can effectively protect the user's data privacy and prevent the theft of user data for excessive marketing. The generation and function implementation of key data can be conventional operating means in the field.

[0104] In another embodiment of the present application, the similarity of the target user's picture is determined by comparing the hash values ​​of the tested picture resource and the comparison picture resource:

[0105] Obtain hash value data of the image resource to be tested and the image resource to be compared respectively, analyze the data elements and data structure in the hash value data, and arrange the data elements and data structure in order;

[0106] The data distance between the image resource to be tested and the image resource to be compared is counted according to the arrangement, and the image similarity of the target user is determined based on the threshold of the data distance.

[0107] It should be noted that the embodiment of the present application draws a conclusion by judging the Hamming distance, wherein the Hamming distance is within 8 to conclude that the pictures are the same. The Hamming distance is the number of steps required to transform a set of binary data into another set of data. This value can measure the difference between two pictures. The smaller the Hamming distance, the higher the similarity. The embodiment of the present application works by finding the data structure in a set of random data, by determining the similarity between elements, and by calculating the distance between them and other elements in the element space.

[0108] Please refer to Figure 2 , Figure 2 This application provides a block diagram of a system for identifying and processing user images in a private domain e-commerce platform.

[0109] A second aspect of an embodiment of the present application provides a recognition processing system based on user images in a private domain e-commerce platform, including a memory 21 and a processor 22, wherein the memory 21 includes a recognition processing program based on user images in the private domain e-commerce platform, and when the recognition processing program based on user images in the private domain e-commerce platform is executed by the processor 22, the following steps are implemented:

[0110] Collect all the image resources generated by the target user on platform A, and decompose and pre-process the collected image resources to obtain thumbnail images;

[0111] Calculate the hash value of the thumbnail image to generate batch hash value data, convert the batch hash value data into batch interval values ​​according to the image information, and store the batch hash value data and the batch interval values ​​in the cloud database;

[0112] Obtain the image resources to be tested generated by the target user on platform B, calculate the hash value data to be tested of the image and generate the corresponding interval value to be tested;

[0113] The numerical value of the interval to be tested is used to search for the associated comparison picture resources in the cloud database, and the similarity of the target user's pictures is determined by comparing the hash values ​​of the picture resources to be tested and the comparison picture resources.

[0114] According to the embodiment of the present application, the collected image resources are decomposed and preprocessed specifically as follows:

[0115] Calculate the RGB color component values ​​of the image resource and determine whether it exceeds the threshold;

[0116] When the RGB color component value exceeds the threshold, the data size of the image resource is adjusted to uniformly limit the RGB color components of the image resource to within the threshold.

[0117] According to an embodiment of the present application, it also includes:

[0118] Obtain the log information generated by the target user under the platform operation node, and filter the data packets with graphic tags in the log information, where the operation nodes include upload, download and forward;

[0119] The data resources are cleaned according to the source information, process information and security information in the data packet to obtain the image resources of the target user in the platform.

[0120] According to the embodiment of the present application, the hash value of the thumbnail image is calculated to generate batch hash value data as follows:

[0121] Get the RGB value of each pixel in the thumbnail image, where the information of the second highest (16-23) red component is taken out to get the r value, the information of the middle (8-15) green component is taken out to get the g value, and the information of the low (0-7) blue component is taken out to get the b value;

[0122] The grayscale value of the image is calculated according to the RGB value of the image, and the RGB value and grayscale value of the image are used to generate the hash value data of the image using the difference hash algorithm.

[0123] According to an embodiment of the present application, searching for the associated comparison picture resources in the cloud database using the value of the interval to be tested is specifically as follows:

[0124] Obtain static information of target users, analyze preference information and application information of target users, and set different weight ratios for values ​​in different intervals according to preference information and application information;

[0125] Retrieve interval data and compare image resources whose relevance is still within the interval range after introducing different weight calculations.

[0126] According to an embodiment of the present application, the batch hash value data is converted into batch interval values ​​according to the image information as follows:

[0127] Partition the hash value data according to the attribute information and function information in the image information, and convert the hash value data into a numerical value in the range of 0 to 32;

[0128] Multiple intervals are created for interval values ​​according to a preset allocation method, and the intervals are allocated to different system nodes.

[0129] According to the embodiment of the present application, the batch hash value data and the batch interval value data are stored in the cloud database specifically as follows:

[0130] Send batch interval values ​​to the db and introduce index labels to them;

[0131] Store the batch hash value data in redis and establish a mapping relationship with the batch interval value.

[0132] According to the embodiment of the present application, the batch hash value data and the batch interval value data are stored in the cloud database specifically as follows:

[0133] Collect digital information of batch interval values, format the digital information and generate key data;

[0134] Use key data to encrypt redis when storing batch hash value data and batch interval value data.

[0135] In another embodiment of the present application, the similarity of the target user's picture is determined by comparing the hash values ​​of the tested picture resource and the comparison picture resource:

[0136] Obtain hash value data of the image resource to be tested and the image resource to be compared respectively, analyze the data elements and data structure in the hash value data, and arrange the data elements and data structure in order;

[0137] The data distance between the image resource to be tested and the image resource to be compared is counted according to the arrangement, and the image similarity of the target user is determined based on the threshold of the data distance.

[0138] A third aspect of an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium includes a recognition processing program based on user images in a private domain e-commerce platform. When the recognition processing program based on user images in a private domain e-commerce platform is executed by a processor, the steps of the recognition processing method based on user images in a private domain e-commerce platform are implemented. For details, see Figure 1 The description of the method steps will not be repeated here.

[0139] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

[0140] The units described above as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units; they may be located in one place or distributed on multiple network units; some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0141] In addition, all functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may be separately used as a unit, or two or more units may be integrated into one unit; the above-mentioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0142] Those skilled in the art can understand that: all or part of the steps of implementing the above method embodiments can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps of the above method embodiments; and the aforementioned storage medium includes: mobile storage devices, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), disks or optical disks, and other media that can store program codes.

[0143] Alternatively, if the above-mentioned integrated unit of the present invention is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present invention can be essentially or partly reflected in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, ROM, RAM, magnetic disks or optical disks.

[0144] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for identifying and processing user images in a private domain e-commerce platform, characterized in that: The steps include: Collect all the image resources generated by the target user on platform A, and decompose and pre-process the collected image resources to obtain thumbnail images; Calculate the hash value of the thumbnail image to generate batch hash value data, convert the batch hash value data into batch interval values ​​according to the image information, and store the batch hash value data and the batch interval values ​​in the cloud database; Obtain the image resources to be tested generated by the target user on platform B, calculate the hash value data to be tested of the image and generate the corresponding interval value to be tested; Use the value of the interval to be tested to search for the associated comparison picture resources in the cloud database, and determine the similarity of the target user's pictures by comparing the hash values ​​of the picture resources to be tested and the comparison picture resources; The hash value of the thumbnail image is calculated to generate batch hash value data as follows: Get the RGB value of each pixel in the thumbnail image, where the information of the second highest (16-23) red component is taken to get the r value, the information of the middle (8-15) green component is taken to get the g value, and the information of the low (0-7) blue component is taken to get the b value; The gray value of the image is calculated according to the RGB value of the image, and the RGB value and the gray value of the image are used to generate the hash value data of the image using the difference hash algorithm; The specific method of searching for the relevant comparison image resources in the cloud database using the values ​​of the interval to be tested is as follows: Obtain static information of target users, analyze preference information and application information of target users, and set different weight ratios for values ​​in different intervals according to preference information and application information; Retrieve the comparative image resources whose relevance is still within the interval after introducing different weight calculations; According to the image information, the batch hash value data is converted into batch interval values ​​as follows: Partition the hash value data according to the attribute information and function information in the image information, and convert the hash value data into a numerical value in the range of 0 to 32; Multiple intervals are created for interval values ​​according to a preset allocation method, and the intervals are allocated to different system nodes.

2. The method for identifying and processing user images in a private domain e-commerce platform according to claim 1 is characterized in that: The collected image resources are decomposed and preprocessed as follows: Calculate the RGB color component values ​​of the image resource and determine whether it exceeds the threshold; When the RGB color component value exceeds the threshold, the data size of the image resource is adjusted to uniformly limit the RGB color components of the image resource to within the threshold.

3. The method for identifying and processing user images in a private domain e-commerce platform according to claim 1 is characterized in that: Also includes: Obtain the log information generated by the target user under the platform operation node, and filter the data packets with graphic tags in the log information, where the operation nodes include upload, download and forward; The data resources are cleaned according to the source information, process information and security information in the data packet to obtain the image resources of the target user in the platform.

4. The method for identifying and processing user images in a private domain e-commerce platform according to claim 1 is characterized in that: The specific steps of storing batch hash value data and batch interval values ​​into the cloud database are as follows: Send batch interval values ​​to the db and introduce index labels to them; Store the batch hash value data in redis and establish a mapping relationship with the batch interval value.

5. The method for identifying and processing user images in a private domain e-commerce platform according to claim 1, characterized in that: The specific steps of storing batch hash value data and batch interval values ​​into the cloud database are as follows: Collect digital information of batch interval values, format the digital information and generate key data; Use key data to encrypt redis when storing batch hash value data and batch interval value data.

6. A recognition and processing system based on user pictures in a private domain e-commerce platform, characterized in that: It includes a memory and a processor, the memory includes a recognition processing program based on user pictures in a private domain e-commerce platform, and when the recognition processing program based on user pictures in the private domain e-commerce platform is executed by the processor, the steps of the recognition processing method based on user pictures in the private domain e-commerce platform as described in any one of claims 1 to 5 are implemented.

7. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a recognition processing program based on user images in a private domain e-commerce platform. When the recognition processing program based on user images in a private domain e-commerce platform is executed by a processor, the steps of the recognition processing method based on user images in a private domain e-commerce platform as described in any one of claims 1 to 5 are implemented.

Images