Picture processing method and device, computer equipment and storage medium

Abstract

The invention belongs to the technical field of artificial intelligence, and relates to a picture processing method and device, computer equipment and a storage medium, and the method comprises the steps: obtaining pictures sent by all customer service staff in a preset time period; filtering all the pictures based on the white list library and the black list library to obtain a specified picture; clustering the specified picture based on a target clustering algorithm to obtain a clustering result; screening out a target cluster from all clusters based on a sorting screening strategy, and carrying out recognition processing on the target cluster based on a picture recognition model to obtain a picture label result; updating a white list library and a black list library based on the picture tag result to obtain a target white list library and a target black list library; and performing violation detection on the target picture sent by the target customer service based on the target white list library and the target black list library to obtain a corresponding violation detection result. In addition, the violation detection result can be stored in the block chain. According to the invention, the processing efficiency of customer service quality inspection processing is improved, and the accuracy of the obtained violation detection result is ensured.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence development technology and financial technology, and in particular to image processing methods, devices, computer equipment and storage media. Background Art

[0002] In the insurance service industry, customer service, as an important bridge between enterprises and consumers, undertakes multiple functions such as information transmission, problem answering, and complaint handling. With the rapid development of information technology and the popularization of Internet insurance products, the working methods of insurance customer service have expanded from traditional telephone services to multi-channel interactions such as online customer service, social media, and instant messaging tools, which has led to a sharp increase in the amount of information exchange and data processing involved in customer service work.

[0003] In the daily operation of insurance customer service, image data is widely used as an intuitive and efficient way of information transmission in multiple links such as answering customer inquiries, submitting claims information, and displaying insurance products. However, this convenience is also accompanied by potential risks and challenges. On the one hand, the pictures sent by customer service when communicating with customers may contain sensitive information, misleading content, or information that does not meet regulatory requirements. If they are released to the public without effective review, they may harm customer interests, trigger a crisis of trust, and even cause the company to face legal risks. On the other hand, due to the complexity and professionalism of the insurance business, the information in the pictures often needs to be accurately interpreted and judged for compliance, which places extremely high demands on the professionalism and rigor of quality inspection work.

[0004] The traditional insurance customer service quality inspection model mainly relies on manual review, that is, the quality inspectors need to review the conversation records between customer service and customers one by one, especially the image data involved, to determine whether they comply with corporate specifications, industry standards and legal and regulatory requirements. This model is not only inefficient and difficult to cope with the needs of massive data processing, but also easily affected by human factors, resulting in subjectivity and inconsistency in the review results, making it difficult to ensure the accuracy of customer service quality inspection. Summary of the invention

[0005] The purpose of the embodiments of the present application is to propose an image processing method, device, computer equipment and storage medium to solve the technical problems that the existing insurance customer service quality inspection model is inefficient and difficult to ensure the accuracy of customer service quality inspection.

[0006] In order to solve the above technical problems, the present application provides an image processing method, which adopts the following technical solutions:

[0007] Get all pictures sent by customer service within a preset time period;

[0008] Filter all the images based on the preset whitelist library and blacklist library to obtain corresponding designated images;

[0009] Performing clustering processing on the specified image based on a preset target clustering algorithm to obtain a corresponding clustering result; wherein the clustering result includes a plurality of clusters;

[0010] Filtering a target cluster from all the clusters based on a preset sorting and filtering strategy; wherein the number of the target clusters includes multiple;

[0011] Based on a preset image recognition model, the target cluster is identified and processed to obtain a corresponding image label result;

[0012] Based on the image label result, the whitelist library and the blacklist library are updated to obtain corresponding target whitelist library and target blacklist library;

[0013] Based on the target whitelist library and the target blacklist library, violation detection processing is performed on the target image sent by the target customer service to obtain a violation detection result corresponding to the target image.

[0014] Furthermore, the step of filtering all the images based on the preset whitelist library and blacklist library to obtain the corresponding designated images specifically includes:

[0015] Performing whitelist matching on all the pictures respectively with the whitelist library, so as to filter out a first picture that passes the whitelist matching from all the pictures;

[0016] Obtaining a second picture from all the pictures; wherein the second picture is other pictures among all the pictures except the first picture;

[0017] Performing blacklist matching on all the second images respectively with the blacklist library, so as to filter out third images that pass the blacklist matching from all the second images;

[0018] filtering the third picture in the second picture to obtain a corresponding fourth picture;

[0019] The fourth picture is used as the designated picture.

[0020] Furthermore, the step of selecting a target cluster from all the clusters based on a preset sorting and screening strategy specifically includes:

[0021] Get the number of images of all the clusters;

[0022] Sorting all the clusters in descending order of the number of pictures to obtain corresponding sorting results;

[0023] Get the specified number of presets;

[0024] Extracting the designated clusters corresponding to the designated number from the sorting results in sequence;

[0025] The designated cluster is used as the target cluster.

[0026] Further, the step of updating the whitelist library and the blacklist library based on the image tag result to obtain the corresponding target whitelist library and target blacklist library specifically includes:

[0027] Based on the image labeling results, specifying compliant images and specifying illegal images that meet the update conditions are screened out from all the target clusters;

[0028] Add the specified compliant image to the whitelist library to obtain the corresponding target whitelist library;

[0029] The designated illegal images are added to the blacklist library to obtain a corresponding target blacklist library.

[0030] Further, the step of performing violation detection processing on the target image sent by the target customer service based on the target whitelist library and the target blacklist library to obtain a violation detection result corresponding to the target image specifically includes:

[0031] Preprocessing the target image to obtain a corresponding fifth image;

[0032] Performing a whitelist comparison between the fifth image and the target whitelist library, and determining whether there is a sixth image in the target whitelist library whose matching degree with the fifth image is greater than a preset first matching threshold;

[0033] If the sixth picture exists in the target whitelist library, generating a first violation detection result that the target picture is a compliant picture;

[0034] If the sixth image does not exist in the target whitelist library, performing a blacklist comparison between the fifth image and the target blacklist library to determine whether there is a seventh image in the target blacklist library whose matching degree with the fifth image is greater than a preset second matching threshold;

[0035] If the seventh picture exists in the target blacklist library, a second violation detection result indicating that the target picture is a violation picture is generated.

[0036] Furthermore, before the step of performing recognition processing on the target cluster based on the preset image recognition model to obtain the corresponding image label result, it also includes:

[0037] Obtain the initial image data after annotation processing;

[0038] Preprocessing the initial image data to obtain corresponding image sample data;

[0039] Call the preset classification model;

[0040] Training and optimizing the classification model based on the image sample data to obtain a specified classification model that meets preset construction conditions;

[0041] The specified classification model is used as the image recognition model.

[0042] Further, after the step of performing violation detection processing on the target image sent by the target customer service based on the target whitelist library and the target blacklist library to obtain a violation detection result corresponding to the target image, the method further includes:

[0043] Determine whether the content of the violation detection result indicates that the target image is a violation image;

[0044] If yes, obtain the preset violation handling strategy;

[0045] Based on the violation handling strategy, corresponding warning processing is performed on the target customer service.

[0046] In order to solve the above technical problems, the embodiment of the present application further provides an image processing device, which adopts the following technical solution:

[0047] The first acquisition module is used to acquire all pictures sent by customer service within a preset time period;

[0048] A filtering module, used to filter all the images based on a preset whitelist library and a blacklist library to obtain corresponding designated images;

[0049] A clustering module, used to perform clustering processing on the specified image based on a preset target clustering algorithm to obtain a corresponding clustering result; wherein the clustering result includes multiple clusters;

[0050] A screening module, used for screening out a target cluster from all the clusters based on a preset sorting and screening strategy; wherein the number of the target clusters includes a plurality;

[0051] A recognition module, used to perform recognition processing on the target cluster based on a preset image recognition model to obtain a corresponding image label result;

[0052] An updating module, used for updating the whitelist library and the blacklist library based on the image label result to obtain a corresponding target whitelist library and a target blacklist library;

[0053] The detection module is used to perform violation detection processing on the target image sent by the target customer service based on the target whitelist library and the target blacklist library to obtain a violation detection result corresponding to the target image.

[0054] In order to solve the above technical problems, the embodiment of the present application further provides a computer device, which adopts the following technical solution:

[0055] Get all pictures sent by customer service within a preset time period;

[0056] Filter all the images based on the preset whitelist library and blacklist library to obtain corresponding designated images;

[0057] Performing clustering processing on the specified image based on a preset target clustering algorithm to obtain a corresponding clustering result; wherein the clustering result includes a plurality of clusters;

[0058] Filtering a target cluster from all the clusters based on a preset sorting and filtering strategy; wherein the number of the target clusters includes multiple;

[0059] Based on a preset image recognition model, the target cluster is identified and processed to obtain a corresponding image label result;

[0060] Based on the image label result, the whitelist library and the blacklist library are updated to obtain corresponding target whitelist library and target blacklist library;

[0061] Based on the target whitelist library and the target blacklist library, violation detection processing is performed on the target image sent by the target customer service to obtain a violation detection result corresponding to the target image.

[0062] In order to solve the above technical problems, the embodiment of the present application further provides a computer-readable storage medium, which adopts the following technical solution:

[0063] Get all pictures sent by customer service within a preset time period;

[0064] Filter all the images based on the preset whitelist library and blacklist library to obtain corresponding designated images;

[0065] Performing clustering processing on the specified image based on a preset target clustering algorithm to obtain a corresponding clustering result; wherein the clustering result includes a plurality of clusters;

[0066] Filtering a target cluster from all the clusters based on a preset sorting and filtering strategy; wherein the number of the target clusters includes multiple;

[0067] Based on a preset image recognition model, the target cluster is identified and processed to obtain a corresponding image label result;

[0068] Based on the image label result, the whitelist library and the blacklist library are updated to obtain corresponding target whitelist library and target blacklist library;

[0069] Based on the target whitelist library and the target blacklist library, violation detection processing is performed on the target image sent by the target customer service to obtain a violation detection result corresponding to the target image.

[0070] Compared with the prior art, the embodiments of the present application have the following beneficial effects:

[0071] The present application first obtains all pictures sent by customer services within a preset time period; then, based on the preset whitelist library and blacklist library, all the pictures are filtered and processed to obtain corresponding designated pictures; then, based on the preset target clustering algorithm, the designated pictures are clustered to obtain corresponding clustering results; wherein, the clustering results include multiple clusters; and based on the preset sorting and screening strategy, a target cluster is screened out from all the clusters; wherein, the number of the target clusters includes multiple; subsequently, the target cluster is identified and processed based on the preset picture recognition model to obtain corresponding picture labeling results; further, based on the picture labeling results, the whitelist library and the blacklist library are updated to obtain corresponding target whitelist library and target blacklist library; finally, based on the target whitelist library and the target blacklist library, the target picture sent by the target customer service is subjected to violation detection processing to obtain a violation detection result corresponding to the target picture. The present application obtains all pictures sent by customer services within a preset time period, and filters all pictures based on the use of a whitelist library and a blacklist library to obtain corresponding designated pictures, then clusters the designated pictures based on the use of a target clustering algorithm to obtain corresponding clustering results, and selects target clusters from all clusters based on the use of a sorting and filtering strategy, and then identifies the target cluster based on the use of an image recognition model to obtain corresponding image label results, and updates the whitelist library and the blacklist library with the image label results to obtain corresponding target whitelist library and target blacklist library, and subsequently, by using the obtained target whitelist library and target blacklist library, it is possible to quickly and accurately complete the violation detection processing of the target pictures sent by the target customer service, effectively improving the processing efficiency of the customer service quality inspection processing, and ensuring the accuracy of the violation detection results of the obtained target pictures. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] In order to more clearly illustrate the scheme in the present application, a brief introduction is given below to the drawings required for use in the description of the embodiments of the present application. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0073] Figure 1 is an exemplary system architecture diagram to which the present application may be applied;

[0074] Figure 2 A flowchart of an embodiment of a method for processing an image according to the present application;

[0075] Figure 3 is a structural schematic diagram of an embodiment of a picture processing device according to the present application;

[0076] Figure 4 It is a structural diagram of an embodiment of a computer device according to the present application. DETAILED DESCRIPTION

[0077] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by technicians in the technical field of the present application; the terms used in the specification of the application herein are only for the purpose of describing specific embodiments and are not intended to limit the present application; the terms "including" and "having" and any variations thereof in the specification and claims of the present application and the above-mentioned drawings are intended to cover non-exclusive inclusions. The terms "first", "second", etc. in the specification and claims of the present application or the above-mentioned drawings are used to distinguish different objects, not to describe a specific order.

[0078] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0079] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings.

[0080] like Figure 1 As shown, the system architecture 100 may include a terminal device 101, a network 102 and a server 103. The terminal device 101 may be a laptop 1011, a tablet computer 1012 or a mobile phone 1013. The network 102 is used to provide a medium for a communication link between the terminal device 101 and the server 103. The network 102 may include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0081] The user can use the terminal device 101 to interact with the server 103 through the network 102 to receive or send messages, etc. Various communication client applications can be installed on the terminal device 101, such as web browser applications, shopping applications, search applications, instant messaging tools, email clients, social platform software, etc.

[0082] The terminal device 101 can be any electronic device with a display screen and supporting web browsing. In addition to the laptop computer 1011, tablet computer 1012 or mobile phone 1013, the terminal device 101 can also be an e-book reader, an MP3 player (Moving Picture Experts Group Audio Layer III), an MP4 (Moving Picture Experts Group Audio Layer IV), a laptop computer, a desktop computer, etc.

[0083] The server 103 may be a server that provides various services, such as a background server that provides support for a web page displayed on the terminal device 101 .

[0084] It should be noted that the image processing method provided in the embodiment of the present application is generally executed by a server / terminal device, and accordingly, the image processing device is generally arranged in the server / terminal device.

[0085] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.

[0086] Continue to refer Figure 2 , shows a flow chart of an embodiment of the image processing method according to the present application. According to different requirements, the order of the steps in the flow chart can be changed, and some steps can be omitted. The image processing method provided in the embodiment of the present application can be applied to any scenario that requires image processing, and the image processing method can be applied to products in these scenarios, for example, image violation detection in the financial insurance field. The image processing method comprises the following steps:

[0087] Step S201, obtaining all pictures sent by customer service within a preset time period.

[0088] In this embodiment, the electronic device (eg, Figure 1The server / terminal device shown in the figure) can obtain the image to be checked for duplicates through a wired connection or a wireless connection. It should be pointed out that the above-mentioned wireless connection method may include but is not limited to 3G / 4G / 5G connection, WiFi connection, Bluetooth connection, WiMAX connection, Zigbee connection, UWB (ultra wideband) connection, and other wireless connection methods currently known or to be developed in the future. The executor of the present application may specifically be an image processing system, or simply referred to as a system. The above-mentioned preset time period may be one week. The pictures sent by all customer services within the above-mentioned preset time period may refer to all picture data sent by customer services in financial enterprises during business exchanges collected every week. In the business scenarios in the financial field, the above-mentioned financial enterprises may include insurance companies, banks, and the like.

[0089] Step S202: Filter all the images based on the preset whitelist library and blacklist library to obtain corresponding designated images.

[0090] In this embodiment, the specific implementation process of filtering all the images based on the preset whitelist library and blacklist library to obtain the corresponding specified images will be further described in detail in subsequent specific embodiments of the present application and will not be elaborated here.

[0091] Step S203, clustering the designated image based on a preset target clustering algorithm to obtain a corresponding clustering result; wherein the clustering result includes a plurality of clusters.

[0092] In this embodiment, the target clustering algorithm may specifically adopt a clustering algorithm such as K-Means and DBSCAN. The target clustering algorithm is used to perform clustering processing on the specified images so as to divide the specified images into different clusters, each cluster representing a group of images with similar features. The target clustering algorithm is used to group the specified images after filtering the images sent by all customer services within the preset time period, so that a large number of images can be automatically classified and sorted. This reduces the workload of manual processing and speeds up the processing.

[0093] Step S204: Filter out a target cluster from all the clusters based on a preset sorting and filtering strategy; wherein the number of the target clusters includes a plurality.

[0094] In this embodiment, the specific implementation process of selecting the target cluster from all the clusters based on the preset sorting and screening strategy will be described in further detail in subsequent specific embodiments of the present application, and will not be elaborated on here.

[0095] Step S205 , performing recognition processing on the target cluster based on a preset image recognition model to obtain a corresponding image label result.

[0096] In this embodiment, by inputting the target cluster into the image recognition model, the image recognition model predicts the images in the target cluster and outputs the label of the target cluster, that is, the image label result. By labeling the target cluster, the content or theme of each cluster can be better understood, which can facilitate subsequent search, screening and management, and improve the overall understanding and utilization of image data. In addition, the model construction process of the above-mentioned image recognition model will be further described in detail in the subsequent specific embodiments of this application, and will not be elaborated on here.

[0097] Step S206: update the whitelist library and the blacklist library based on the image label result to obtain a corresponding target whitelist library and a target blacklist library.

[0098] In this embodiment, the above-mentioned specific implementation process of updating the whitelist library and the blacklist library based on the image label result to obtain the corresponding target whitelist library and the target blacklist library will be further described in detail in the subsequent specific embodiments of this application, and will not be elaborated here.

[0099] Step S207: Perform violation detection processing on the target image sent by the target customer service based on the target whitelist library and the target blacklist library to obtain a violation detection result corresponding to the target image.

[0100] In the present embodiment, the above-mentioned specific implementation process of performing violation detection processing on the target image sent by the target customer service based on the target whitelist library and the target blacklist library to obtain the violation detection result corresponding to the target image will be further described in detail in the subsequent specific embodiments of the present application and will not be elaborated in detail here.

[0101] The present application first obtains all pictures sent by customer services within a preset time period; then, based on the preset whitelist library and blacklist library, all the pictures are filtered and processed to obtain corresponding designated pictures; then, based on the preset target clustering algorithm, the designated pictures are clustered to obtain corresponding clustering results; wherein, the clustering results include multiple clusters; and based on the preset sorting and screening strategy, a target cluster is screened out from all the clusters; wherein, the number of the target clusters includes multiple; subsequently, the target cluster is identified and processed based on the preset picture recognition model to obtain corresponding picture labeling results; further, based on the picture labeling results, the whitelist library and the blacklist library are updated to obtain corresponding target whitelist library and target blacklist library; finally, based on the target whitelist library and the target blacklist library, the target picture sent by the target customer service is subjected to violation detection processing to obtain a violation detection result corresponding to the target picture. The present application obtains all pictures sent by customer services within a preset time period, and filters all pictures based on the use of a whitelist library and a blacklist library to obtain corresponding designated pictures, then clusters the designated pictures based on the use of a target clustering algorithm to obtain corresponding clustering results, and selects target clusters from all clusters based on the use of a sorting and filtering strategy, and then identifies the target cluster based on the use of an image recognition model to obtain corresponding image label results, and updates the whitelist library and the blacklist library with the image label results to obtain corresponding target whitelist library and target blacklist library, and subsequently, by using the obtained target whitelist library and target blacklist library, it is possible to quickly and accurately complete the violation detection processing of the target pictures sent by the target customer service, effectively improving the processing efficiency of the customer service quality inspection processing, and ensuring the accuracy of the violation detection results of the obtained target pictures.

[0102] In some optional implementations, step S202 includes the following steps:

[0103] All the pictures are respectively matched against the whitelist library to select a first picture that passes the whitelist match from all the pictures.

[0104] In this embodiment, each collected image can be matched and compared with the images in the above whitelist library by using image recognition technology (such as hash comparison, feature extraction comparison, etc.). If the match is successful (that is, the similarity between the images exceeds a certain threshold), the image is regarded as a compliant image, directly released and recorded as the first image.

[0105] Obtain a second picture from all the pictures; wherein the second picture is other pictures among all the pictures except the first picture.

[0106] In this embodiment, the first picture may be filtered from all pictures to obtain the corresponding second picture, wherein the second picture refers to a picture that does not pass the whitelist matching.

[0107] Blacklist matching is performed on all the second pictures respectively with the blacklist library to filter out third pictures that pass the blacklist matching from all the second pictures.

[0108] In this embodiment, the same image recognition technology is used to match and compare each second image with the images in the blacklist library. If the match is successful, it is regarded as an illegal image and used as the third image that passes the blacklist match.

[0109] The third picture is filtered in the second picture to obtain a corresponding fourth picture.

[0110] In this embodiment, the fourth picture is obtained by filtering out the selected third picture from the second picture, and is used as the final designated picture.

[0111] The fourth picture is used as the designated picture.

[0112] In this embodiment, by using the whitelist library and the blacklist library, known valid images and invalid or bad content can be quickly filtered out from all the images sent by the customer service, thereby improving the filtering accuracy of all the images sent by the customer service. This can save time and resources for subsequent processing, so that the focus can be on processing the remaining image data obtained after filtering the images sent by all the customer service, that is, the specified images.

[0113] The present application performs whitelist matching on all the pictures respectively with the whitelist library to filter out the first picture that passes the whitelist matching from all the pictures; then obtains the second picture from all the pictures; wherein the second picture is the other pictures in all the pictures except the first picture; then performs blacklist matching on all the second pictures respectively with the blacklist library to filter out the third picture that passes the blacklist matching from all the second pictures; subsequently, the third picture is filtered in the second pictures to obtain the corresponding fourth picture, and the fourth picture is used as the designated picture. The present application performs filtering processing on all pictures based on the use of the whitelist library and the blacklist library to efficiently and accurately obtain the designated pictures that do not pass the whitelist matching and the blacklist matching, thereby improving the efficiency of obtaining the designated pictures and ensuring the accuracy of the designated pictures obtained. In addition, only the target clustering algorithm needs to be used to cluster the designated pictures, without clustering all the pictures, which effectively reduces the processing workload of picture processing.

[0114] In some optional implementations of this embodiment, step S204 includes the following steps:

[0115] Get the number of images of all the clusters.

[0116] In this embodiment, the number of images in a cluster can be used as a metric for cluster screening. The more images a cluster contains, the larger it is, which means that these clusters occupy a relatively large proportion in the data set and are therefore more likely to contain important information or new violation types.

[0117] All the clusters are sorted in descending order of the number of pictures to obtain corresponding sorting results.

[0118] In this embodiment, the clusters may be sorted in descending order according to the number of pictures in the clusters to obtain a corresponding sorting result.

[0119] Gets the specified number of presets.

[0120] In this embodiment, the above-mentioned specified number refers to the number of clusters that need to be screened. There is no specific limitation on the selection of the value of the specified number, which can be set according to actual business needs, for example, it can be set to 20.

[0121] Specified clusters corresponding to the specified number are extracted in sequence from the sorting results.

[0122] In this embodiment, starting from the cluster ranked first in the above sorting result, designated clusters corresponding to the designated number may be extracted in sequence and used as the final target clusters.

[0123] The designated cluster is used as the target cluster.

[0124] The present application obtains the number of images in all the clusters; then sorts all the clusters in order of the number of images from large to small to obtain the corresponding sorting result; then obtains the preset specified number; and extracts the specified clusters corresponding to the preset specified number from the sorting result in sequence; and subsequently uses the specified cluster as the target cluster. The present application obtains the number of images in all the clusters, and then sorts all the clusters in order of the number of images from large to small to obtain the corresponding sorting result, and then extracts the specified clusters corresponding to the preset specified number from the sorting result in sequence, so that the desired target cluster can be quickly and accurately screened out from all the clusters, thereby improving the screening efficiency and screening accuracy of the target cluster.

[0125] In some optional implementations, step S206 includes the following steps:

[0126] Based on the image labeling results, designated compliant images and designated noncompliant images that meet the update conditions are screened out from all the target clusters.

[0127] In this embodiment, content analysis can be performed on the above-mentioned image labeling results to screen out designated compliant images and designated illegal images that meet the update conditions from all target clusters. The designated compliant images that meet the update conditions refer to images corresponding to the newly discovered compliant image types. The designated illegal images that meet the update conditions refer to images corresponding to the newly discovered illegal image types. In addition, the new compliant image types refer to image types corresponding to normal behaviors that were not previously included in the whitelist. The new illegal image types refer to image types corresponding to fraudulent behaviors that were not previously included in the whitelist.

[0128] The specified compliant image is added to the whitelist library to obtain the corresponding target whitelist library.

[0129] In this embodiment, the above-mentioned designated compliant images are added to the whitelist library to complete the update process of the whitelist library, thereby obtaining the corresponding target whitelist library.

[0130] The designated illegal images are added to the blacklist library to obtain a corresponding target blacklist library.

[0131] In this embodiment, the above-mentioned designated illegal pictures are added to the blacklist library to complete the update process of the blacklist library, thereby obtaining the corresponding target blacklist library. By using the designated compliant pictures and designated illegal pictures that meet the update conditions selected from all target clusters to update and optimize the whitelist library and the blacklist library respectively, the target whitelist library and the target blacklist library that can be obtained can accurately reflect the new illegal pictures and compliant pictures.

[0132] In addition, by updating the blacklist and whitelist libraries, the filtering system can be continuously optimized to adapt to the characteristics of newly emerging illegal and compliant images. Over time, the accuracy and adaptability of the system will be improved.

[0133] In addition, you can increase the counter and switch to the collection and processing of customer service image data for the next time period. Through iterative processing, repeat the above steps of image processing until all weeks of image data are processed. And over time, continuously optimize and adjust the processing flow to improve the accuracy and adaptability of the system. Through iterative operations, the system can adapt to image data from different weeks and continuously improve the processing effect. This iterative processing can gradually accumulate experience and adjust and optimize the process.

[0134] This application, based on the image labeling results, screens out designated compliant images and designated illegal images that meet the update conditions from all the target clusters; then adds the designated compliant images to the whitelist library to obtain the corresponding target whitelist library; and subsequently adds the designated illegal images to the blacklist library to obtain the corresponding target blacklist library. This application, based on the image labeling results, screens out designated compliant images and designated illegal images that meet the update conditions from all target clusters, then adds the designated compliant images to the whitelist library to obtain the corresponding target whitelist library, and adds the designated illegal images to the blacklist library to obtain the corresponding target blacklist library, thereby achieving efficient and accurate update processing of the whitelist library and the blacklist library, and effectively improving the data accuracy of the generated target whitelist library and the target blacklist library. This is conducive to the subsequent use of the target whitelist library and the target blacklist library to accurately complete the violation detection processing of the target images sent by the customer service.

[0135] In some optional implementations, step S207 includes the following steps:

[0136] The target image is preprocessed to obtain a corresponding fifth image.

[0137] In this embodiment, after receiving the target image sent by the target customer service, the target image may be preprocessed, such as denoising, grayscale (if applicable), size adjustment, etc., to ensure the image quality of the generated fifth image.

[0138] The fifth image is compared with the target whitelist library to determine whether there is a sixth image in the target whitelist library whose matching degree with the fifth image is greater than a preset first matching threshold.

[0139] In this embodiment, the fifth image can be compared with the target whitelist library by using image recognition technology (such as hash comparison, feature extraction comparison, etc.) to obtain a corresponding matching result. There is no specific limitation on the value of the first matching threshold, which can be set according to actual business needs.

[0140] If the sixth picture exists in the target whitelist library, a first violation detection result indicating that the target picture is a compliant picture is generated.

[0141] In this embodiment, if it is detected that there is a picture in the target whitelist library whose matching degree with the fifth picture is greater than the first matching threshold, the fifth picture is determined to be compliant, and a first violation detection result indicating that the target picture is a compliant picture is generated.

[0142] If the sixth picture does not exist in the target whitelist library, a blacklist comparison is performed on the fifth picture and the target blacklist library to determine whether there is a seventh picture in the target blacklist library whose matching degree with the fifth picture is greater than a preset second matching threshold.

[0143] In this embodiment, the fifth image can be compared with the target blacklist library by using the above image recognition technology to obtain a corresponding matching result. There is no specific limitation on the value of the second matching threshold, which can be set according to actual business needs.

[0144] If the seventh picture exists in the target blacklist library, a second violation detection result indicating that the target picture is a violation picture is generated.

[0145] In this embodiment, if there is a picture in the target blacklist library whose matching degree with the fifth picture is greater than the second matching threshold, the fifth picture is determined to be illegal, and a second violation detection result indicating that the target picture is an illegal picture is generated.

[0146] The present application obtains the corresponding fifth image by preprocessing the target image; then performs a whitelist comparison on the fifth image and the target whitelist library to determine whether there is a sixth image in the target whitelist library whose matching degree with the fifth image is greater than a preset first matching threshold; if the sixth image exists in the target whitelist library, a first violation detection result is generated that the target image is a compliant image; and if the sixth image does not exist in the target whitelist library, the fifth image is compared with the target blacklist library to determine whether there is a seventh image in the target blacklist library whose matching degree with the fifth image is greater than a preset second matching threshold; if the seventh image exists in the target blacklist library, a second violation detection result is generated that the target image is a violation image. The present application performs a list comparison on the target image sent by the target customer service based on the use of the target whitelist library and the target blacklist library, and then according to the obtained list comparison result, the violation detection processing of the target image can be completed quickly and accurately and the corresponding violation detection result can be generated, effectively ensuring the accuracy of the generated violation detection result.

[0147] In some optional implementations of this embodiment, before step S205, the electronic device may further perform the following steps:

[0148] Get the initial image data after annotation.

[0149] In this embodiment, initial image data that has been manually reviewed and labeled during customer service communication is collected in advance.

[0150] The initial image data is preprocessed to obtain corresponding image sample data.

[0151] In this embodiment, the above preprocessing includes preprocessing the initial image data, including adjusting the image size, normalizing pixel values, etc., to obtain corresponding image sample data, wherein the image sample data will be used as a data set for training the classification model.

[0152] Call the preset classification model.

[0153] In this embodiment, the above classification model can specifically adopt a convolutional neural network structure, including an input layer, a convolution layer, a pooling layer, a fully connected layer and an output layer. Among them, the appropriate network depth and width can be selected according to the complexity of the problem and the data scale. In addition, a classification loss function (such as cross entropy loss) corresponding to the classification model is selected, and an optimizer (such as Adam, SGD, etc.) is defined to update the network weights.

[0154] The classification model is trained and optimized based on the image sample data to obtain a specified classification model that meets the preset construction conditions.

[0155] In this embodiment, the preprocessed image sample data is input into the classification model, forward propagation and back propagation are performed, and the network weights are continuously updated by the optimizer until a preset number of training rounds is reached or a stopping condition is met, thereby obtaining a corresponding specified classification model.

[0156] The specified classification model is used as the image recognition model.

[0157] In this embodiment, after the construction of the image recognition model is completed, the model parameters of the specified classification model are saved so that the model parameters of the specified classification model can be loaded for prediction in subsequent use.

[0158] The present application obtains the initial image data after the annotated processing; then pre-processes the initial image data to obtain the corresponding image sample data; then calls the preset classification model; and trains and optimizes the classification model based on the image sample data to obtain a specified classification model that meets the preset construction conditions; and subsequently uses the specified classification model as the image recognition model. The present application obtains the image sample data after the annotated processing by pre-processing the initial image data; then trains and optimizes the classification model based on the image sample data to achieve the rapid and accurate construction of the image recognition model that meets the preset construction conditions, thereby improving the construction efficiency of the image recognition model and ensuring the model effect of the obtained image recognition model.

[0159] In some optional implementations of this embodiment, after step S207, the electronic device may further perform the following steps:

[0160] Determine whether the content of the violation detection result is that the target image is a violation image.

[0161] In this embodiment, the content of the violation detection result may include that the target image is a violation image, or that the target image is a compliance image.

[0162] If yes, get the preset violation handling strategy.

[0163] In this embodiment, the policy content of the above-mentioned violation handling policy includes: restricting the client from continuing to handle the user's request or question, notifying the customer service and taking corresponding punishment measures.

[0164] Based on the violation handling strategy, corresponding warning processing is performed on the target customer service.

[0165] In this embodiment, early warning processing can be performed on the target customer service according to the policy content of the above-mentioned violation processing policy.

[0166] This application determines whether the content of the violation detection result indicates that the target image is a violation image; if so, obtains a preset violation handling strategy; and subsequently performs corresponding early warning processing on the target customer service based on the violation handling strategy. When this application detects that the content of the violation detection result indicates that the target image is a violation image, it will automatically and intelligently perform corresponding early warning processing on the target customer service based on the use of the violation handling strategy, thereby effectively ensuring the compliance of the customer service behavior.

[0167] In some optional implementations, the user information obtained is subject to the user's consent and complies with relevant laws and policies.

[0168] In addition, the non-Company software tools or components that appear in the embodiments of the present application are merely examples and do not represent actual use.

[0169] It should be understood that the order of execution of the steps in the above embodiment does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.

[0170] It should be emphasized that in order to further ensure the privacy and security of the above-mentioned violation detection results, the above-mentioned violation detection results can also be stored in a node of a blockchain.

[0171] The blockchain referred to in this application is a new application model of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanism, encryption algorithm, etc. Blockchain is essentially a decentralized database, a string of data blocks generated by cryptographic methods. Each data block contains a batch of network transaction information, which is used to verify the validity of its information (anti-counterfeiting) and generate the next block. Blockchain can include the underlying blockchain platform, platform product service layer, and application service layer.

[0172] The embodiments of the present application can acquire and process relevant data based on artificial intelligence technology. Among them, artificial intelligence (AI) is the theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.

[0173] AI basic technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing technology, operation / interaction systems, mechatronics, etc. AI software technologies mainly include computer vision technology, robotics technology, biometrics technology, speech processing technology, natural language processing technology, and machine learning / deep learning.

[0174] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through computer-readable instructions, and the computer-readable instructions can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, the aforementioned storage medium can be a non-volatile storage medium such as a disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

[0175] It should be understood that, although the steps in the flowchart of the accompanying drawings are displayed in sequence as indicated by the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least a part of the steps in the flowchart of the accompanying drawings may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be executed in turn or alternately with other steps or at least a part of the sub-steps or stages of other steps.

[0176] Further references Figure 3 , as a response to the above Figure 2 The present application provides an embodiment of a picture processing device, and the embodiment of the device is Figure 2 Corresponding to the method embodiment shown, the device can be specifically applied to various electronic devices.

[0177] like Figure 3 As shown, the image processing device 300 described in this embodiment includes: a first acquisition module 301, a filtering module 302, a clustering module 303, a screening module 304, a recognition module 305, an updating module 306 and a detection module 307. Among them:

[0178] The first acquisition module 301 is used to acquire all pictures sent by customer service within a preset time period;

[0179] The filtering module 302 is used to filter all the pictures based on the preset whitelist library and blacklist library to obtain the corresponding designated pictures;

[0180] The clustering module 303 is used to perform clustering processing on the specified image based on a preset target clustering algorithm to obtain a corresponding clustering result; wherein the clustering result includes multiple clusters;

[0181] A screening module 304, configured to screen out a target cluster from all the clusters based on a preset sorting and screening strategy; wherein the number of the target clusters includes a plurality;

[0182] The recognition module 305 is used to perform recognition processing on the target cluster based on a preset image recognition model to obtain a corresponding image label result;

[0183] An updating module 306 is used to update the whitelist library and the blacklist library based on the image label result to obtain a corresponding target whitelist library and a target blacklist library;

[0184] The detection module 307 is used to perform violation detection processing on the target image sent by the target customer service based on the target whitelist library and the target blacklist library to obtain a violation detection result corresponding to the target image.

[0185] In this embodiment, the operations performed by the above modules or units correspond one-to-one to the steps of the image processing method in the aforementioned implementation mode, and are not described in detail here.

[0186] In some optional implementations of this embodiment, the filtering module 302 includes:

[0187] A first matching submodule, configured to perform whitelist matching on all the images respectively with the whitelist library, so as to filter out a first image that passes the whitelist matching from all the images;

[0188] A first acquisition submodule is used to acquire a second picture from all the pictures; wherein the second picture is other pictures among all the pictures except the first picture;

[0189] A second matching submodule, configured to perform blacklist matching on all the second images respectively with the blacklist library, so as to filter out third images that pass the blacklist matching from all the second images;

[0190] A filtering submodule, configured to filter the third image in the second image to obtain a corresponding fourth image;

[0191] The first determining submodule is configured to use the fourth picture as the designated picture.

[0192] In this embodiment, the operations performed by the above modules or units correspond one-to-one to the steps of the image processing method in the aforementioned implementation mode, and are not described in detail here.

[0193] In some optional implementations of this embodiment, the screening module 304 includes:

[0194] A second acquisition submodule is used to obtain the number of pictures of all the clusters;

[0195] A sorting submodule is used to sort all the clusters in descending order of the number of pictures to obtain corresponding sorting results;

[0196] The third acquisition submodule is used to acquire a preset specified quantity;

[0197] An extraction submodule, used to extract the specified clusters corresponding to the specified number from the sorting results in sequence;

[0198] The second determining submodule is configured to use the designated cluster as the target cluster.

[0199] In this embodiment, the operations performed by the above modules or units correspond one-to-one to the steps of the image processing method in the aforementioned implementation mode, and are not described in detail here.

[0200] In some optional implementations of this embodiment, the updating module 306 includes:

[0201] A screening submodule, configured to screen out designated compliant images and designated illegal images that meet the update conditions from all the target clusters based on the image label results;

[0202] A first adding submodule is used to add the specified compliant image to the whitelist library to obtain a corresponding target whitelist library;

[0203] The second adding submodule is used to add the designated illegal image to the blacklist library to obtain a corresponding target blacklist library.

[0204] In this embodiment, the operations performed by the above modules or units correspond one-to-one to the steps of the image processing method in the aforementioned implementation mode, and are not described in detail here.

[0205] In some optional implementations of this embodiment, the detection module 307 includes:

[0206] A preprocessing submodule, used to preprocess the target image to obtain a corresponding fifth image;

[0207] A first judgment submodule is used to perform a whitelist comparison between the fifth image and the target whitelist library, and determine whether there is a sixth image in the target whitelist library whose matching degree with the fifth image is greater than a preset first matching threshold;

[0208] A first generating submodule, configured to generate a first violation detection result that the target image is a compliant image if the sixth image exists in the target whitelist library;

[0209] a second judgment submodule, configured to, if the sixth image does not exist in the target whitelist library, perform a blacklist comparison between the fifth image and the target blacklist library to determine whether there is a seventh image in the target blacklist library whose matching degree with the fifth image is greater than a preset second matching threshold;

[0210] The second generating module is configured to generate a second violation detection result indicating that the target image is a violation image if the seventh image exists in the target blacklist library.

[0211] In this embodiment, the operations performed by the above modules or units correspond one-to-one to the steps of the image processing method in the aforementioned implementation mode, and are not described in detail here.

[0212] In some optional implementations of this embodiment, the image processing device further includes:

[0213] The second acquisition module is used to acquire the initial image data after the annotation processing;

[0214] A preprocessing module, used to preprocess the initial image data to obtain corresponding image sample data;

[0215] A calling module is used to call a preset classification model;

[0216] A processing module, used to train and optimize the classification model based on the image sample data to obtain a specified classification model that meets the preset construction conditions;

[0217] A determination module is used to use the specified classification model as the image recognition model.

[0218] In this embodiment, the operations performed by the above modules or units correspond one-to-one to the steps of the image processing method in the aforementioned implementation mode, and are not described in detail here.

[0219] In some optional implementations of this embodiment, the image processing device further includes:

[0220] A judgment module, used for judging whether the content of the violation detection result indicates that the target image is a violation image;

[0221] The third acquisition module is used to obtain a preset violation handling strategy if yes;

[0222] The early warning module is used to perform corresponding early warning processing on the target customer service based on the violation processing strategy.

[0223] In this embodiment, the operations performed by the above modules or units correspond one-to-one to the steps of the image processing method in the aforementioned implementation mode, and are not described in detail here.

[0224] To solve the above technical problems, the present application also provides a computer device. Figure 4 , Figure 4 This is a basic structural block diagram of the computer device in this embodiment.

[0225] The computer device 4 includes a memory 41, a processor 42, and a network interface 43 that are interconnected through a system bus. It should be noted that the figure only shows a computer device 4 with components 41-43, but it should be understood that it is not required to implement all the components shown, and more or fewer components can be implemented instead. Among them, those skilled in the art can understand that the computer device here is a device that can automatically perform numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes but is not limited to microprocessors, application specific integrated circuits (Application Specific Integrated Circuit, ASIC), programmable gate arrays (Field-Programmable Gate Array, FPGA), digital processors (Digital Signal Processor, DSP), embedded devices, etc.

[0226] The computer device may be a computing device such as a desktop computer, a notebook, a PDA, a cloud server, etc. The computer device may interact with a user through a keyboard, a mouse, a remote controller, a touch pad, or a voice control device.

[0227] The memory 41 includes at least one type of readable storage medium, and the readable storage medium includes flash memory, hard disk, multimedia card, card-type memory (for example, SD or DX memory, etc.), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, disk, optical disk, etc. In some embodiments, the memory 41 can be an internal storage unit of the computer device 4, such as a hard disk or memory of the computer device 4. In other embodiments, the memory 41 can also be an external storage device of the computer device 4, such as a plug-in hard disk equipped on the computer device 4, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (FlashCard), etc. Of course, the memory 41 can also include both the internal storage unit of the computer device 4 and its external storage device. In this embodiment, the memory 41 is generally used to store the operating system and various application software installed on the computer device 4, such as computer-readable instructions of the image processing method, etc. In addition, the memory 41 can also be used to temporarily store various types of data that have been output or are to be output.

[0228] The processor 42 may be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chips in some embodiments. The processor 42 is generally used to control the overall operation of the computer device 4. In this embodiment, the processor 42 is used to run computer-readable instructions stored in the memory 41 or process data, such as computer-readable instructions for running the image processing method.

[0229] The network interface 43 may include a wireless network interface or a wired network interface. The network interface 43 is generally used to establish a communication connection between the computer device 4 and other electronic devices.

[0230] Compared with the prior art, the embodiments of the present application have the following beneficial effects:

[0231] In an embodiment of the present application, by obtaining all pictures sent by customer services within a preset time period, and filtering all pictures based on the use of a whitelist library and a blacklist library to obtain corresponding designated pictures, then clustering the designated pictures based on the use of a target clustering algorithm to obtain corresponding clustering results, and filtering out target clusters from all clusters based on the use of a sorting and filtering strategy, and then identifying the target cluster based on the use of an image recognition model to obtain corresponding image label results, and updating the whitelist library and the blacklist library based on the image label results to obtain corresponding target whitelist library and target blacklist library, and subsequently using the obtained target whitelist library and target blacklist library, it is possible to quickly and accurately complete the violation detection processing of the target pictures sent by the target customer service, effectively improving the processing efficiency of the customer service quality inspection processing, and ensuring the accuracy of the violation detection results of the obtained target pictures.

[0232] The present application also provides another implementation, namely, providing a computer-readable storage medium, wherein the computer-readable storage medium stores computer-readable instructions, and the computer-readable instructions can be executed by at least one processor to enable the at least one processor to perform the steps of the image processing method as described above.

[0233] Compared with the prior art, the embodiments of the present application have the following beneficial effects:

[0234] In an embodiment of the present application, by obtaining all pictures sent by customer services within a preset time period, and filtering all pictures based on the use of a whitelist library and a blacklist library to obtain corresponding designated pictures, then clustering the designated pictures based on the use of a target clustering algorithm to obtain corresponding clustering results, and filtering out target clusters from all clusters based on the use of a sorting and filtering strategy, and then identifying the target cluster based on the use of an image recognition model to obtain corresponding image label results, and updating the whitelist library and the blacklist library based on the image label results to obtain corresponding target whitelist library and target blacklist library, and subsequently using the obtained target whitelist library and target blacklist library, it is possible to quickly and accurately complete the violation detection processing of the target pictures sent by the target customer service, effectively improving the processing efficiency of the customer service quality inspection processing, and ensuring the accuracy of the violation detection results of the obtained target pictures.

[0235] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in each embodiment of the present application.

[0236] Obviously, the embodiments described above are only some embodiments of the present application, rather than all embodiments. The preferred embodiments of the present application are given in the accompanying drawings, but they do not limit the patent scope of the present application. The present application can be implemented in many different forms. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosure of the present application more thorough and comprehensive. Although the present application is described in detail with reference to the aforementioned embodiments, for those skilled in the art, it is still possible to modify the technical solutions recorded in the aforementioned specific implementation methods, or to perform equivalent replacement of some of the technical features therein. Any equivalent structure made using the contents of the specification and drawings of this application, directly or indirectly used in other related technical fields, is similarly within the scope of patent protection of this application.

Claims

1. A method for processing an image, characterized in that: The steps include: Get all pictures sent by customer service within a preset time period; Filter all the images based on the preset whitelist library and blacklist library to obtain corresponding designated images; Performing clustering processing on the specified image based on a preset target clustering algorithm to obtain a corresponding clustering result; wherein the clustering result includes a plurality of clusters; Filtering a target cluster from all the clusters based on a preset sorting and filtering strategy; wherein the number of the target clusters includes multiple; Based on a preset image recognition model, the target cluster is identified and processed to obtain a corresponding image label result; Based on the image label result, the whitelist library and the blacklist library are updated to obtain corresponding target whitelist library and target blacklist library; Based on the target whitelist library and the target blacklist library, violation detection processing is performed on the target image sent by the target customer service to obtain a violation detection result corresponding to the target image.

2. The image processing method according to claim 1, characterized in that: The step of filtering all the images based on the preset whitelist library and blacklist library to obtain the corresponding designated images specifically includes: Performing whitelist matching on all the pictures respectively with the whitelist library, so as to filter out a first picture that passes the whitelist matching from all the pictures; Obtaining a second picture from all the pictures; wherein the second picture is other pictures among all the pictures except the first picture; Performing blacklist matching on all the second images respectively with the blacklist library, so as to filter out third images that pass the blacklist matching from all the second images; filtering the third picture in the second picture to obtain a corresponding fourth picture; The fourth picture is used as the designated picture.

3. The image processing method according to claim 1, characterized in that: The step of selecting a target cluster from all the clusters based on a preset sorting and screening strategy specifically includes: Get the number of images of all the clusters; Sorting all the clusters in descending order of the number of pictures to obtain corresponding sorting results; Get the specified number of presets; Extracting the designated clusters corresponding to the designated number from the sorting results in sequence; The designated cluster is used as the target cluster.

4. The image processing method according to claim 1, characterized in that: The step of updating the whitelist library and the blacklist library based on the image tag result to obtain the corresponding target whitelist library and target blacklist library specifically includes: Based on the image labeling results, specifying compliant images and specifying illegal images that meet the update conditions are screened out from all the target clusters; Add the specified compliant image to the whitelist library to obtain the corresponding target whitelist library; The designated illegal images are added to the blacklist library to obtain a corresponding target blacklist library.

5. The image processing method according to claim 1, characterized in that: The step of performing violation detection processing on the target image sent by the target customer service based on the target whitelist library and the target blacklist library to obtain a violation detection result corresponding to the target image specifically includes: Preprocessing the target image to obtain a corresponding fifth image; Performing a whitelist comparison between the fifth image and the target whitelist library, and determining whether there is a sixth image in the target whitelist library whose matching degree with the fifth image is greater than a preset first matching threshold; If the sixth picture exists in the target whitelist library, generating a first violation detection result that the target picture is a compliant picture; If the sixth image does not exist in the target whitelist library, performing a blacklist comparison between the fifth image and the target blacklist library to determine whether there is a seventh image in the target blacklist library whose matching degree with the fifth image is greater than a preset second matching threshold; If the seventh picture exists in the target blacklist library, a second violation detection result indicating that the target picture is a violation picture is generated.

6. The image processing method according to claim 1, characterized in that: Before the step of performing recognition processing on the target cluster based on the preset image recognition model to obtain the corresponding image label result, the method further includes: Obtain the initial image data after annotation processing; Preprocessing the initial image data to obtain corresponding image sample data; Call the preset classification model; Training and optimizing the classification model based on the image sample data to obtain a specified classification model that meets preset construction conditions; The specified classification model is used as the image recognition model.

7. The image processing method according to claim 1, characterized in that: After the step of performing violation detection processing on the target image sent by the target customer service based on the target whitelist library and the target blacklist library to obtain a violation detection result corresponding to the target image, the method further includes: Determine whether the content of the violation detection result indicates that the target image is a violation image; If yes, obtain the preset violation handling strategy; Based on the violation handling strategy, corresponding warning processing is performed on the target customer service.

8. A picture processing device, characterized in that: include: The first acquisition module is used to acquire all pictures sent by customer service within a preset time period; A filtering module, used to filter all the images based on a preset whitelist library and a blacklist library to obtain corresponding designated images; A clustering module, used to perform clustering processing on the specified image based on a preset target clustering algorithm to obtain a corresponding clustering result; wherein the clustering result includes multiple clusters; A screening module, used for screening out a target cluster from all the clusters based on a preset sorting and screening strategy; wherein the number of the target clusters includes a plurality; A recognition module, used to perform recognition processing on the target cluster based on a preset image recognition model to obtain a corresponding image label result; An updating module, used for updating the whitelist library and the blacklist library based on the image label result to obtain a corresponding target whitelist library and a target blacklist library; The detection module is used to perform violation detection processing on the target image sent by the target customer service based on the target whitelist library and the target blacklist library to obtain a violation detection result corresponding to the target image.

9. A computer device, characterized in that: The method comprises a memory and a processor, wherein the memory stores computer-readable instructions, and the processor implements the steps of the image processing method according to any one of claims 1 to 7 when executing the computer-readable instructions.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-readable instructions, and when the computer-readable instructions are executed by a processor, the steps of the image processing method according to any one of claims 1 to 7 are implemented.

Images