Privacy information processing method and device, electronic equipment and medium

Abstract

The invention discloses a privacy information processing method and device, electronic equipment and a medium, and belongs to the technical field of information processing. The method comprises the steps of determining at least one target privacy operation corresponding to each privacy information area based on privacy information content corresponding to each privacy information area in a to-be-processed picture and text content associated with the to-be-processed picture; and executing a corresponding target privacy operation on each privacy information area in the to-be-processed picture.

Description

Technical Field

[0001] This application belongs to the field of information processing technology, and specifically relates to a method, apparatus, electronic device and medium for processing privacy information. Background Technology

[0002] With the widespread use of electronic devices, image files have become an important means of daily communication. However, image files may contain sensitive personal information, such as ID card numbers, bank card numbers, geolocation information, and facial images. The leakage of this information can pose serious privacy risks. Existing privacy processing technologies typically rely on static rules to apply fixed privacy treatments to different types of information. This can lead to over- or under-protection of some privacy information, increasing the risk of user privacy breaches. Summary of the Invention

[0003] The purpose of this application is to provide a method, apparatus, electronic device, and medium for processing privacy information, which can avoid the problem that using the same privacy operations on different privacy information content can lead to overprotection or leakage risks for some privacy information content.

[0004] In a first aspect, embodiments of this application provide a method for processing privacy information, including: Based on the privacy information content corresponding to each privacy information region in the image to be processed and the text content associated with the image to be processed, at least one target privacy operation corresponding to each privacy information region is determined. Perform corresponding target privacy operations on each privacy information region in the image to be processed.

[0005] Secondly, embodiments of this application provide a privacy information processing apparatus, including: The determination module is used to determine at least one target privacy operation corresponding to each privacy information region based on the privacy information content corresponding to each privacy information region in the image to be processed and the text content associated with the image to be processed. The execution module is used to perform corresponding target privacy operations on each privacy information region in the image to be processed.

[0006] Thirdly, embodiments of this application provide an electronic device including a processor and a memory, wherein the memory stores programs or instructions executable on the processor, and the programs or instructions, when executed by the processor, implement the steps of the method described in the first aspect.

[0007] Fourthly, embodiments of this application provide a readable storage medium on which a program or instructions are stored, which, when executed by a processor, implement the steps of the method described in the first aspect.

[0008] Fifthly, embodiments of this application provide a chip, the chip including a processor and a communication interface, the communication interface being coupled to the processor, the processor being used to run programs or instructions to implement the method as described in the first aspect.

[0009] In a sixth aspect, embodiments of this application provide a computer program product stored in a storage medium, which is executed by at least one processor to implement the method described in the first aspect.

[0010] In this embodiment, firstly, based on the privacy information content corresponding to each privacy information region in the image to be processed and the text content associated with the image, at least one target privacy operation is determined for each privacy information region. The corresponding target privacy operation is then executed on each privacy information region in the image. That is, by distinguishing different privacy information content through the text content associated with the image and the privacy information content of each privacy information region, different target privacy operations are determined for different privacy information content. This not only avoids the problem of overprotecting or leaking some privacy information content due to using the same target privacy operation for different privacy information content, but also improves the accuracy of selecting appropriate privacy operations and increases the security of privacy information. Furthermore, the entire process is automated, allowing users to achieve high security of privacy information while maintaining ease of operation, greatly reducing the number of user intervention steps. Attached Figure Description

[0011] Figure 1 This is a flowchart illustrating a method for processing privacy information provided in some embodiments of this application; Figure 2 This is a schematic diagram illustrating the selection of images to be processed provided in some embodiments of this application; Figure 3 This is a schematic flowchart illustrating the privacy information processing method provided in some embodiments of this application; Figure 4 These are schematic diagrams of scene labels provided in some embodiments of this application; Figure 5 This is a schematic diagram illustrating the process of determining the target privacy processing strategy provided in some embodiments of this application; Figure 6 These are schematic diagrams illustrating the structured privacy tags provided in some embodiments of this application; Figure 7 This is a schematic diagram illustrating the selection operation of structured privacy tags provided in some embodiments of this application; Figure 8 This is a schematic diagram of the structure of a privacy information processing device provided in some embodiments of this application; Figure 9 These are structural block diagrams of electronic devices provided in some embodiments of this application; Figure 10 These are structural block diagrams of electronic devices provided in some embodiments of this application. Detailed Implementation

[0012] The technical solutions of the embodiments of this application will be clearly described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application are within the scope of protection of this application.

[0013] The terms "first," "second," etc., used in the specification and claims of this application are used to distinguish similar objects and not to describe a specific order or sequence. It should be understood that such terms can be used interchangeably where appropriate so that embodiments of this application can be implemented in orders other than those illustrated or described herein, and the objects distinguished by "first," "second," etc., are generally of the same class and the number of objects is not limited; for example, a first object can be one or more. Furthermore, in the specification and claims, "and / or" indicates at least one of the connected objects, and the character " / " generally indicates that the preceding and following objects are in an "or" relationship.

[0014] Current privacy information identification technologies have the following drawbacks: 1. Insufficient ability to identify privacy information: Existing technologies mainly rely on image recognition models and optical character recognition (OCR) technology. However, due to complex backgrounds, low image quality, and different formats of privacy information, the recognition accuracy of these methods is low, making it difficult to fully identify and protect all potential privacy information.

[0015] 2. Lack of scenario adaptability: Current technologies typically perform only fixed privacy processing after identifying privacy information, without adjusting privacy processing strategies according to the different requirements of actual scenarios (such as social media, instant messaging, work emails, etc.). Therefore, there are issues of over- or under-protection of privacy.

[0016] 3. Insufficiently intelligent privacy processing: Existing privacy processing technologies are usually based on static rules, which are difficult to dynamically adjust according to user needs, leading to an increased risk of user privacy leakage and a poor user experience.

[0017] 4. Excessive manual intervention: Although some methods allow users to manually select privacy processing policies, this method is cumbersome and prone to errors, and cannot achieve real-time and comprehensive privacy protection.

[0018] Therefore, embodiments of this application provide a method, apparatus, electronic device, and medium for processing privacy information. These methods employ different privacy processing strategies for different categories of privacy information, avoiding the problem of over-protection or leakage risks associated with using the same privacy processing strategy for different categories. Furthermore, after determining the privacy processing strategy, the different privacy information is further refined and differentiated based on specific privacy categories and content, thereby increasing the security of the privacy information. The entire process is automated, allowing users to achieve high security while maintaining ease of operation, significantly reducing the number of user intervention steps.

[0019] The method for processing privacy information provided in this application will be described below with reference to the accompanying drawings, through specific embodiments and application scenarios.

[0020] like Figure 1 As shown in the figure, this application provides a method for processing privacy information, which may specifically include the following steps: Step 101: Based on the privacy information content corresponding to each privacy information region in the image to be processed and the text content associated with the image to be processed, determine at least one target privacy operation corresponding to each privacy information region.

[0021] Specifically, the user selects one or more images from a plurality of images, and uses the selected images as the images to be processed, for example: Figure 2 As shown, the images to be processed are Image 1, Image 2, and Image 3. For each image to be processed, an automatic extraction operation is performed on the privacy information in the image to be processed, obtaining the privacy information content and the corresponding privacy information region. This region can be a rectangle or other shape that surrounds the privacy information. When a user needs to share the image to be processed, they can not only select the image to be processed but also enter text content associated with the image. Based on the privacy information content corresponding to each privacy information region in the image to be processed and the text content associated with the image, at least one target privacy operation is determined for each privacy information region. That is, by using the text content associated with the image to be processed and the privacy information content of each privacy information region, different privacy information content is distinguished, and different target privacy operations are determined for different privacy information content. This not only avoids the problem of overprotecting or leaking some privacy information content by using the same target privacy operation for different privacy information content, but also improves the accuracy of selecting appropriate privacy operations and increases the security of privacy information.

[0022] Step 102: Perform corresponding target privacy operations on each privacy information region in the image to be processed.

[0023] Specifically, after determining the target privacy operation corresponding to each privacy information area, processing instructions for each privacy information area are generated based on the target privacy operation and processing instruction template. The processing instructions for each privacy information area are then summarized to obtain a processing instruction set. The image processing engine is called through the processing instruction set to execute the target privacy operation corresponding to each privacy information area according to the processing instruction set, generate the target image after the privacy operation, and share the target image and associated text content to the target platform. At the same time, sharing logs and processing reports are recorded to facilitate subsequent traceability and feedback optimization.

[0024] It should be noted that the electronic device assigns a unique identifier, instruction_set_id, to each generated processing instruction set for subsequent tracking, backtracking, and log auditing, while also recording the identifier field of the associated image to be processed.

[0025] In this embodiment, firstly, based on the privacy information content corresponding to each privacy information region in the image to be processed and the text content associated with the image, at least one target privacy operation is determined for each privacy information region. The corresponding target privacy operation is then executed on each privacy information region in the image. That is, by distinguishing different privacy information content through the text content associated with the image and the privacy information content of each privacy information region, different target privacy operations are determined for different privacy information content. This not only avoids the problem of overprotecting or leaking some privacy information content due to using the same target privacy operation for different privacy information content, but also improves the accuracy of selecting appropriate privacy operations and increases the security of privacy information. Furthermore, the entire process is automated, allowing users to achieve high security of privacy information while maintaining ease of operation, greatly reducing the number of user intervention steps.

[0026] In one optional embodiment, step 101 determines at least one target privacy operation corresponding to each privacy information region based on the privacy information content corresponding to each privacy information region in the image to be processed and the text content associated with the image to be processed, specifically including: Based on the privacy information content corresponding to each privacy information area and the text content associated with the image to be processed, the privacy risk score of the image to be processed, the confidence level of the privacy information corresponding to each privacy information area, and the privacy category are obtained. Based on the privacy risk score and the confidence level of privacy information corresponding to each privacy information area, the target privacy processing strategy corresponding to each privacy information area is determined. Based on the privacy information content corresponding to each privacy information area and the privacy category, at least one target privacy operation is determined from the multiple privacy operations included in the target privacy processing strategy. Specifically, based on the privacy information content corresponding to each privacy information region in the image to be processed and the text content associated with the image to be processed, the overall privacy risk score of the image to be processed, the confidence level of the privacy information corresponding to each privacy information region, and the privacy category corresponding to each privacy information region are calculated.

[0027] Based on a comprehensive privacy risk score and the confidence level of privacy information corresponding to each privacy information region, the target privacy processing strategy for each privacy information region can be determined. Therefore, from a macro perspective, the privacy risk score is used to comprehensively judge the overall privacy risk of an image, and from a micro perspective, the confidence level of privacy information in each privacy information region is used to evaluate the recognition credibility of each privacy information region. By determining the target privacy processing strategy for each privacy information region from both macro and micro perspectives, we can not only avoid the problem of overprotecting some privacy information or posing a risk of leakage when using the same privacy processing strategy for different privacy information, but also improve the accuracy of selecting appropriate privacy processing strategies.

[0028] Based on the privacy information content and privacy category corresponding to each privacy information area, at least one target privacy operation is determined from the multiple privacy operations included in the predetermined target privacy processing strategy. The corresponding target privacy operation is then executed on each privacy information area in the image to be processed. That is, after determining the target privacy processing strategy, different privacy information is further distinguished by specific privacy categories and privacy information content, thereby further increasing the security of privacy information.

[0029] Each target privacy processing strategy includes multiple privacy operations, such as deletion, replacement, occlusion, and blurring. Different target privacy processing strategies contain different privacy operations, and some privacy operations may be shared among different target privacy processing strategies.

[0030] like Figure 3 As shown below, the process of processing the aforementioned privacy information will be explained through a specific embodiment: Step 301: The user selects the image to be processed from multiple images and obtains the text content associated with the image to be processed.

[0031] Step 302: Using visual recognition and metadata parsing methods, extract the image information located by each method and perform semantic analysis on the text content.

[0032] Step 303: By extracting image information and performing semantic analysis on the text content, identify sensitive entities in the image to be processed and generate a privacy information list containing multiple privacy information.

[0033] Step 304: Obtain the privacy information content and coordinate information of each privacy information, and determine the corresponding privacy information area based on the privacy information content and coordinate information.

[0034] Step 305: Based on the privacy information content corresponding to each privacy information area and the text content associated with the image to be processed, obtain the privacy risk score of the image to be processed, the confidence level of the privacy information corresponding to each privacy information area, and the privacy category.

[0035] Step 306: Based on the privacy risk score and the confidence level of privacy information corresponding to each privacy information area, determine the target privacy processing strategy for each privacy information area.

[0036] Step 307: Based on the privacy information content and privacy category corresponding to each privacy information area, determine at least one target privacy operation from the multiple privacy operations included in the target privacy processing strategy.

[0037] Step 308: Generate processing instructions for each privacy information area based on the target privacy operation and processing instruction template corresponding to each privacy information area.

[0038] Step 309: Summarize the processing instructions for each privacy information area to obtain a processing instruction set.

[0039] Step 310: Call the image processing engine through the processing instruction set, so that the image processing engine can execute the target privacy operations corresponding to each privacy information area according to the processing instruction set, and generate the target image after privacy operations.

[0040] Step 311: Share the target image and associated text content to the target platform. In the above embodiment, firstly, based on the privacy information content corresponding to each privacy information region in the image to be processed and the text content associated with the image to be processed, obtain the comprehensive privacy risk score of the image to be processed, the privacy information confidence level corresponding to each privacy information region, and the privacy category corresponding to each privacy information region. Based on the comprehensive privacy risk score and the privacy information confidence level corresponding to each privacy information region, the target privacy processing strategy corresponding to each privacy information region can be determined. Therefore, from a macro perspective, the privacy risk score is used to comprehensively judge the overall privacy risk of the image, and from a micro perspective, the privacy confidence level of each privacy information region is used to evaluate the recognition credibility of each privacy information region. Determining the target privacy processing strategy corresponding to each privacy information region from both macro and micro perspectives can not only avoid the problem of some privacy information being overprotected or at risk of leakage due to the same privacy processing strategy for different privacy information, but also improve the accuracy of selecting appropriate privacy processing strategies.

[0041] Furthermore, based on the privacy information content and category corresponding to each privacy information area, at least one target privacy operation is determined from multiple privacy operations included in the target privacy processing strategy. This target privacy operation is then executed on each privacy information area in the image to be processed. In other words, after determining the target privacy processing strategy, different privacy information is further refined and differentiated based on specific privacy categories and content, thereby increasing the security of privacy information. The entire process is automated, allowing users to achieve high privacy security while maintaining ease of operation, greatly reducing the number of user intervention steps.

[0042] In an optional specific embodiment, the step of obtaining the privacy risk score of the image to be processed based on the privacy information content corresponding to each privacy information area and the text content associated with the image to be processed specifically includes steps 401 to 405: Step 401: Based on the privacy information content corresponding to each privacy information area and the text content associated with the image to be processed, obtain multi-dimensional structured data content, and perform vectorization encoding processing on the structured data content of each dimension to obtain the feature vector corresponding to the structured data content of each dimension.

[0043] Specifically, based on the template of structured data in multiple dimensions, the structured data content of each dimension is obtained based on the privacy information content corresponding to each privacy information region in the image to be processed and the text content associated with the image to be processed. The structured data content of each dimension is then vectorized and encoded to generate feature vectors corresponding to the structured data content of each dimension.

[0044] For example, there are four dimensions: operational context, semantic intent, environmental security, and content sensitivity. The specific meanings of each dimension are shown in Table 1. Table 1

[0045] In the dimension of operational context, the target application refers to the application currently being used by the user. The recipient type refers to the type of user receiving the images and text content to be processed, such as colleagues or family members. The sharing method refers to the way the user shares the images and text content to be processed, such as internal sharing within an enterprise instant messaging system, posting to social media platforms, or private storage.

[0046] In the dimension of semantic intent, "analyzing text content and identifying sharing purpose and semantic scenario" refers to analyzing text content to identify the user's sharing purpose and semantic scenario, such as: work report, social media post, technical help request, complaint or report.

[0047] In the context of environmental security, the current network status refers to whether the network being used is secure. If it is secure (e.g., an encrypted intranet), the current network status is secure; conversely, if it is not secure (e.g., public Wi-Fi), the current network status is insecure. Device security status refers to whether the electronic device being used is a commonly used or authorized device. If it is not a commonly used or authorized device, the device security status is insecure; conversely, if it is a commonly used or authorized device, the device security status is secure. Assessing external leakage risk refers to evaluating whether there is a risk of privacy leakage based on the current network status and device security status. If both the current network status and device security status are secure, there is no risk of privacy leakage; conversely, if at least one of the current network status or device security status is insecure, there is a risk of privacy leakage, categorized as high, medium, or low risk.

[0048] In the dimension of content sensitivity, image privacy content refers to whether the image to be processed contains privacy information, such as: ID card photos containing privacy information, images of geographical locations containing privacy information, and ordinary images not containing privacy information.

[0049] Step 402: Input the feature vectors corresponding to the structured data content of each dimension into the privacy risk level assessment model corresponding to each dimension to obtain the first privacy risk level score for each dimension.

[0050] Specifically, different dimensions correspond to different privacy risk level assessment models. For any dimension, the feature vector corresponding to the structured data content of that dimension is input into the privacy risk level assessment model corresponding to that dimension. The privacy risk level assessment model processes the feature vector of that dimension to obtain the first privacy risk level score corresponding to that dimension.

[0051] Among them, privacy risk level assessment models include, but are not limited to, the following learning models: logistic regression model, multilayer neural network, convolutional neural network, recurrent neural network, etc.

[0052] Step 403: Based on the first privacy risk level score corresponding to each dimension, determine the mapping relationship between the first privacy risk level score range and the privacy risk level.

[0053] Specifically, multiple dimensions correspond to multiple primary privacy risk level scores, which are then sorted by size. These sorted scores form multiple primary privacy risk level score ranges, with different ranges corresponding to different privacy risk levels, thus establishing a mapping relationship between primary privacy risk level score ranges and privacy risk levels.

[0054] For example: The number of first privacy risk level scores is 4. The 4 first privacy risk level scores, sorted from smallest to largest, are: First Privacy Risk Level Score A, First Privacy Risk Level Score B, First Privacy Risk Level Score C, and First Privacy Risk Level Score D. The score range from 0 to A represents the extremely low privacy risk level, the score range between A and B represents the low privacy risk level, the score range between B and C represents the medium privacy risk level, the score range between C and D represents the high privacy risk level, and the score range greater than D represents the extremely high privacy risk level.

[0055] Step 404: Input the mapping relationship and the structured data content of each dimension into the privacy risk comprehensive assessment model to obtain the scene label, privacy risk level and second privacy risk level score corresponding to each dimension.

[0056] Specifically, the mapping relationship between the first privacy risk level score range and the privacy risk level, as well as the structured data content of each dimension, are input into the privacy risk comprehensive assessment model to obtain at least one scenario label (e.g., sharing scenario) for each dimension, the privacy risk level for each dimension, and the second privacy risk level score for each dimension.

[0057] For example: Figure 4As shown, during the information interaction between the user and the company (Contact Person 1), if the user wants to send an image and text content to be processed, scene tags corresponding to each dimension are obtained based on the image and text content to be processed, such as: scene tag 1, scene tag 2, scene tag 3, ..., scene tag XX.

[0058] Step 405: Calculate the privacy risk score of the image to be processed using a weighted geometric average algorithm based on the second privacy risk level score and weight coefficient corresponding to each scene label.

[0059] Specifically, the privacy risk score of the image to be processed is calculated using a weighted geometric average algorithm based on the second privacy risk level score corresponding to each scene label and the preset weight coefficients corresponding to each scene label.

[0060] For example: the second privacy risk level score corresponding to scenario label A is 5, the second privacy risk level score corresponding to scenario label B is 2, and the second privacy risk level score corresponding to scenario label C is 3. The weight coefficient corresponding to scenario label A is 5, the weight coefficient corresponding to scenario label B is 4, and the weight coefficient corresponding to scenario label C is 1. Therefore, the privacy risk score is S.

[0061] In an optional specific embodiment, the steps of obtaining the privacy information confidence level and privacy category corresponding to each privacy information region based on the privacy information content corresponding to each privacy information region and the text content associated with the image to be processed specifically include: Based on the text content associated with the image to be processed, privacy information is identified in the image to be processed, and multiple privacy information areas and corresponding privacy information content are determined. Based on the privacy information content corresponding to each privacy information area, determine the privacy category corresponding to each privacy information area; The confidence level of privacy information in each privacy information area is calculated using the confidence level calculation method corresponding to the privacy category.

[0062] Specifically, the user selects one or more images from a pool of pictures, and these selected images are designated as images to be processed. For each image, visual recognition and metadata parsing methods are used to extract its localized information. For example, visual recognition extracts watermarks, identity information, and financial information, while metadata parsing extracts the shooting time, location, and relationships between various pieces of information. Based on the extracted information and semantic understanding of the text content, sensitive entities are identified, a privacy information list containing multiple privacy-related details is generated, and the privacy information content and coordinates of each detail are obtained. The corresponding privacy information area is then determined based on the privacy information content and coordinates.

[0063] Based on the privacy information content corresponding to each privacy information area in the image to be processed, the privacy category corresponding to each privacy information area is determined. Privacy categories include, but are not limited to: person feature categories (e.g., face), personal information categories (e.g., ID card number, bank card number), geolocation categories (e.g., geolocation information), and other text categories (e.g., express delivery tracking number).

[0064] Different privacy categories correspond to different confidence calculation methods. The confidence score for each privacy information region is calculated using the confidence calculation method corresponding to its privacy category. For example, for the personal feature category, image processing and visual recognition algorithms are used to calculate the privacy confidence score; for the personal information category, an object detection model is used to generate the privacy confidence score; and for other text categories, optical character recognition (OCR) is used to calculate the privacy confidence score.

[0065] In an optional specific embodiment, the step of determining the target privacy processing strategy corresponding to each privacy information region based on the privacy risk score and the privacy information confidence level corresponding to each privacy information region specifically includes: The privacy information region with a confidence level greater than or equal to the confidence threshold is defined as the first privacy information region; Based on the privacy category corresponding to each first privacy information area, the privacy processing strategy corresponding to the privacy category of each first privacy information area is determined from the first privacy processing strategy set; If the privacy risk score is greater than the risk score threshold, a target privacy processing strategy corresponding to each first privacy information region is determined from the first privacy processing strategy set based on the privacy processing strategy corresponding to each first privacy information region. The priority of the target privacy processing strategy is higher than the priority of the first privacy processing strategy. If the privacy risk score is less than or equal to the risk score threshold, a target privacy processing strategy corresponding to each first privacy information region is determined from the first privacy processing strategy set based on the privacy processing strategy corresponding to each first privacy information region. The priority of the target privacy processing strategy is lower than the priority of the first privacy processing strategy.

[0066] Specifically, the first privacy processing strategy set contains multiple privacy processing strategies, and the contents of the first privacy processing strategy set are shown in Table 2.

[0067] Table 2

[0068] The privacy processing strategies mentioned above are arranged in order of priority from low to high: basic privacy processing strategy, adaptive dynamic strategy, personalized preference strategy, regulatory compliance strategy, and multimodal collaboration strategy.

[0069] First, the privacy information regions with a confidence level greater than or equal to the confidence threshold are defined as the first privacy information regions. For each first privacy information region, based on the privacy category corresponding to each first privacy information region, the privacy processing strategy corresponding to each privacy category is determined from the first privacy processing strategy set. For example, the personal feature category corresponds to an adaptive dynamic strategy (e.g., global blurring when there is high risk), the personal information category corresponds to a basic privacy processing strategy (e.g., blurring or occlusion), and the geographic location category corresponds to a regulatory compliance strategy.

[0070] If the privacy risk score is greater than the risk score threshold (e.g., 0.8, which can be set as needed), the privacy processing strategy corresponding to each first privacy information area will be updated to the target privacy processing strategy with higher privacy priority in the first privacy processing strategy set. That is, in the first privacy processing strategy set, the priority of the target privacy processing strategy is higher than the priority of the first privacy processing strategy. For example, the basic privacy processing strategy will be updated to an adaptive dynamic strategy.

[0071] If the privacy risk score is less than or equal to the risk score threshold, the privacy processing strategy corresponding to each first privacy information area will be updated to the target privacy processing strategy with lower privacy priority in the first privacy processing strategy set. That is, in the first privacy processing strategy set, the priority of the target privacy processing strategy is lower than the priority of the first privacy processing strategy. For example, the adaptive dynamic strategy will be updated to the basic privacy processing strategy.

[0072] In an optional specific embodiment, the step of determining the target privacy processing strategy corresponding to each privacy information region based on the privacy risk score and the privacy information confidence level corresponding to each privacy information region further includes: The privacy information region whose confidence level is less than the confidence threshold is defined as the second privacy information region; Based on the mapping relationship between privacy categories and privacy processing strategies in the second privacy processing strategy set, determine the privacy processing strategy corresponding to the privacy category of each second privacy information area from the second privacy processing strategy set; Based on the coordinate information of each second privacy information region, calculate the intersection-union ratio of every two second privacy information regions; Based on the privacy processing strategy corresponding to each second privacy information region and the intersection-union ratio of every two second privacy information regions, the target privacy processing strategy corresponding to each second privacy information region is determined.

[0073] Specifically, privacy information regions with a confidence level below a certain threshold are defined as second privacy information regions. For these second privacy information regions, the clarity and location of the content are first adjusted. Furthermore, a pre-defined set of second privacy processing strategies is established, containing multiple privacy processing strategies and a mapping relationship between each strategy and a privacy category. The privacy processing strategies in this set are then prioritized. Based on the mapping relationship between privacy categories and the privacy processing strategies in the second privacy processing strategy set, the privacy processing strategy corresponding to each privacy category of the second privacy information region is determined.

[0074] For example: Privacy processing strategy L1: blacking out / encryption, corresponding to the personal information category; Privacy processing strategy L2: high-intensity blurring / mosaic, corresponding to the personal characteristics category; Privacy processing strategy L3: symbolic replacement / low-intensity blurring, corresponding to the geolocation category. The priority order is: L1>L2>L3.

[0075] Furthermore, based on the coordinate information of each second privacy information region, a spatial index tree (such as an R-Tree or quadtree) is constructed to enable rapid retrieval of the spatial relationships of the second privacy information regions. Based on the spatial index tree, all second privacy information regions are traversed, and the intersection-union ratio (IUR) of every pair of second privacy information regions is calculated. Based on the privacy processing strategy corresponding to each second privacy information region and the IUR of every pair of second privacy information regions, the privacy processing strategy corresponding to each second privacy information region is adjusted to determine the target privacy processing strategy for each second privacy information region.

[0076] In an optional specific embodiment, the step of determining the target privacy processing strategy corresponding to each second privacy information region based on the privacy processing strategy corresponding to each second privacy information region and the intersection-union ratio of every two second privacy information regions specifically includes: If the intersection-union ratio of the two second privacy information regions is less than or equal to a preset value, the privacy processing strategy corresponding to each of the two second privacy information regions is determined as the target privacy processing strategy corresponding to each region. If the intersection-union ratio of the two second privacy information regions is greater than the preset value, the conflicting region and the non-conflicting region of the two second privacy information regions are determined. The privacy processing strategy with higher priority among the two privacy processing strategies corresponding to the second privacy information regions is determined as the target privacy processing strategy corresponding to the conflict region. The privacy processing strategies corresponding to the two second privacy information regions are determined as the target privacy processing strategies corresponding to their respective non-conflicting regions.

[0077] Specifically, if the intersection-union ratio of two second privacy information regions is less than or equal to a preset value (e.g., 0), it is determined that there is no spatial conflict between the two second privacy information regions. Then, the privacy processing strategy corresponding to each of the two second privacy information regions is determined as the target privacy processing strategy. That is, the privacy processing strategy corresponding to each of the two second privacy information regions is the target privacy processing strategy.

[0078] If the intersection-union ratio of two second privacy information regions is greater than a preset value (e.g., 0), it is determined that there is a spatial conflict between the two second privacy information regions. A conflict adjacency matrix is ​​then generated to distinguish the conflicting region (i.e., the overlapping region) from the non-conflicting regions (i.e., the non-overlapping regions) within each second privacy information region. Furthermore, based on the proportion of the conflicting region's area, the spatial conflict type can be further divided into three states: complete conflict, partial conflict, and boundary conflict.

[0079] For conflicting regions, if the privacy processing policies corresponding to the two second privacy information regions have different priorities, the privacy processing policy with the higher priority among the two second privacy information regions will be adopted as the target privacy processing policy for that conflicting region. For non-conflicting regions among the two second privacy information regions, the previously determined privacy processing policy will remain as the target privacy processing policy.

[0080] For conflicting regions, the system compares their respective security priorities. Conflicting regions are then forced to adopt the processing strategy corresponding to the privacy entity with the highest priority.

[0081] In an optional specific embodiment, the step of determining the target privacy processing strategy corresponding to each second privacy information region based on the privacy processing strategy corresponding to each second privacy information region and the intersection-union ratio of every two second privacy information regions further includes: If the privacy processing strategies corresponding to the two second privacy information regions have the same priority, the privacy processing strategy corresponding to the privacy operation with higher priority among the privacy operations contained in the privacy processing strategies corresponding to the two second privacy information regions shall be determined as the target privacy processing strategy corresponding to the conflict region.

[0082] Specifically, if the privacy processing policies corresponding to two second privacy information areas have the same priority, the priority of the privacy operations contained in the two privacy processing policies is determined, and the privacy operation with the higher priority is taken as the target privacy processing policy for the conflicting area. If the privacy operations contained in the privacy processing policies corresponding to two second privacy information areas have the same priority, either privacy operation is taken as the target privacy processing policy for the conflicting area.

[0083] For example: If the privacy processing policies corresponding to two second privacy information areas are both privacy processing policy L1, but the privacy operation of one of them is encryption and the privacy operation of the other is blacking out, and the priority of encryption is higher than the priority of blacking out, then encryption is selected as the privacy operation for the conflicting area.

[0084] Alternatively, the degree of different privacy operations can be determined based on the privacy information content in the second privacy information area. Priority is then determined according to the degree of privacy operation; a higher degree of privacy operation results in a higher priority. For example, bank card numbers have a higher priority than ID numbers, therefore, privacy operations on bank card numbers have a higher degree of priority, while those on ID numbers have a lower degree of priority. If both operations involve adjusting opacity, then a higher opacity operation is performed on bank card numbers, and a lower opacity operation is performed on ID numbers.

[0085] like Figure 5 As shown below, a specific embodiment will be used to illustrate the situation where the confidence level of the aforementioned privacy information is less than the confidence threshold: Step 501: Determine the privacy information region where the confidence level of the privacy information is less than the confidence level threshold as the second privacy information region.

[0086] Step 502: Based on the mapping relationship between privacy categories and privacy processing strategies in the second privacy processing strategy set, determine the privacy processing strategy corresponding to the privacy category of each second privacy information area from the second privacy processing strategy set.

[0087] Step 503: Based on the coordinate information of each second privacy information region, calculate the intersection-union ratio of every two second privacy information regions.

[0088] Step 504: If the intersection-union ratio of the two second privacy information regions is less than or equal to a preset value, the privacy processing strategy corresponding to each of the two second privacy information regions is determined as the target privacy processing strategy corresponding to each region.

[0089] Step 505: If the intersection-union ratio of the two second privacy information regions is greater than a preset value, determine the conflicting and non-conflicting regions of the two second privacy information regions.

[0090] Step 506: Determine the privacy processing strategy corresponding to each of the two second privacy information areas as the target privacy processing strategy corresponding to their respective non-conflicting areas.

[0091] Step 507: When the privacy processing strategies corresponding to the two second privacy information areas have different priorities, the privacy processing strategy with higher priority among the two second privacy information areas is determined as the target privacy processing strategy corresponding to the conflict area.

[0092] Step 508: If the privacy processing strategies corresponding to the two second privacy information areas have the same priority, the privacy processing strategy corresponding to the privacy operation with higher priority among the privacy operations contained in the privacy processing strategies of the two second privacy information areas shall be determined as the target privacy processing strategy for the conflict area.

[0093] In the above embodiments, if no privacy processing strategy is matched in the first privacy processing strategy set, the privacy processing strategy corresponding to the privacy category of each second privacy information region is determined from the second privacy processing strategy set. An intersection-union (IUU) is used to determine whether two adjacent second privacy information regions conflict. If a conflict exists, the privacy processing strategy used in the conflicting region is determined based on the priority of the privacy processing strategies. If the privacy processing strategies corresponding to two second privacy information regions have the same priority, the privacy processing strategy used in the conflicting region is further determined based on the priority of privacy operations, thereby maximizing the security of privacy information. In an optional specific embodiment, the method further includes: Based on the coordinate information corresponding to each second privacy information region, the privacy category, and the target privacy processing strategy, the privacy processing strategy is reconstructed and added to the first privacy processing strategy set.

[0094] Specifically, after determining the target privacy processing strategy corresponding to each second privacy information region, the processing method of each second privacy information region is templated to generate a new privacy processing strategy. The strategy intent (e.g., blackening occlusion, superimposed Gaussian blur) is used as the name of the new privacy processing strategy. The coordinate information, privacy category, and privacy operations such as feathering parameters and transparency in the target privacy processing strategy corresponding to each second privacy information region are added to the new privacy processing strategy. The newly constructed privacy processing strategy is added to the first privacy processing strategy set to update the first privacy processing strategy set.

[0095] In an optional specific embodiment, step 102 performs corresponding target privacy operations on each privacy information region in the image to be processed, including: Based on the coordinate information corresponding to each privacy information region and the target privacy operation, the operation parameters of the target privacy operation corresponding to each privacy information region are determined. Based on the operation parameters of the target privacy operation corresponding to each privacy information region, the corresponding target privacy operation is performed on each privacy information region in the image to be processed.

[0096] Specifically, after determining the target privacy operation corresponding to each privacy information region, processing instructions for each privacy information region are generated based on the target privacy operation and the processing instruction template (the content of the processing instruction template is shown in Table 3). The processing instructions include the target privacy operation, the coordinate information of the operation region (i.e., the privacy information region), operation parameters, execution priority, and metadata. The processing instructions for each privacy information region are then summarized to obtain a processing instruction set. Then, according to the priority and dependencies of the privacy processing strategy to which the target privacy operation belongs, the execution of the target privacy operations in each privacy information region is sorted, generating a unified execution order field. Furthermore, the electronic device calculates a checksum and generation time for the processing instruction set for integrity verification.

[0097] Table 3

[0098] Where x1, y1, x2, y2 refer to the coordinates of the privacy information area, radius refers to the radius of 7 determined based on the size of the privacy information area, and blurring is performed based on the radius. color refers to the occlusion color, and a default color can be set as needed. Opacity refers to the transparency of the color, and the transparency is determined according to the privacy category. text refers to the preset text or whitespace fill. Algorithm refers to the encryption algorithm. key_id refers to the encryption key. For example, if the coordinates of the privacy information area are (100, 150, 300, 350) and the radius is 15 millimeters, then Blur(x1, y1, x2, y2, radius) is (100, 150, 300, 350, 15).

[0099] In an optional specific embodiment, after step 104 performs corresponding target privacy operations on each privacy information region in the image to be processed, the method further includes: Based on at least one of the coordinate information corresponding to each privacy information region, the scene label, the privacy category, and the target privacy processing strategy, generate an interactive structured privacy label corresponding to each privacy information region; In response to a triggering operation on a structured privacy label in the privacy information area, at least one of the following is displayed: the coordinate information, the scene label, the privacy category, and the target privacy processing strategy corresponding to the privacy information area.

[0100] Specifically, based on at least one of the following: coordinate information, scene label, privacy category, and target privacy processing strategy corresponding to each privacy information region, an interactive structured privacy label is generated for each privacy information region. For example, such as... Figure 6 As shown, Privacy 1 is a structured privacy label for one privacy information area, and Privacy 2 is a structured privacy label for another privacy information area.

[0101] Structured privacy labels are displayed in the corresponding privacy information areas so that users can intuitively see what privacy information is contained in the image. If a user performs a trigger operation on a structured privacy label in the privacy information area, at least one of the following will be displayed: the coordinates of the privacy information area, the scene label, the privacy category, and the target privacy processing strategy. This trigger operation is the operation of selecting a structured label to view details.

[0102] If a user performs a selection operation on the structured privacy labels in the privacy information area, then the user selects which privacy information areas will implement the corresponding target privacy processing policy and which privacy information areas will not implement the corresponding target privacy processing policy. For example, such as Figure 7 As shown, when a user clicks the structured privacy label for Privacy 1, a "Confirm Cancel?" control is displayed. If the user clicks the "Confirm Cancel?" control, Privacy 1 will be deselected, and the corresponding privacy processing policy will not be applied to the privacy information area corresponding to Privacy 1.

[0103] In summary, the embodiments of this application comprehensively assess the overall privacy risk of an image from a macro perspective using privacy risk scores, and evaluate the recognition credibility of each privacy information region from a micro perspective using the privacy confidence level of each privacy information region. By determining the target privacy processing strategy corresponding to each privacy information region from both macro and micro perspectives, this approach not only avoids the problem of some privacy information being overprotected or at risk of leakage due to the use of the same privacy processing strategy for different privacy information, but also improves the accuracy of selecting appropriate privacy processing strategies.

[0104] Furthermore, based on the privacy information content and category corresponding to each privacy information area, at least one target privacy operation is determined from multiple privacy operations included in the target privacy processing strategy. This target privacy operation is then executed on each privacy information area in the image to be processed. In other words, after determining the target privacy processing strategy, different privacy information is further refined and differentiated through specific privacy categories and content, further enhancing the security of privacy information and ensuring the security of user privacy information during image transmission. The entire process is automated, allowing users to achieve high privacy security while maintaining ease of operation, greatly reducing the number of user intervention steps.

[0105] The privacy information processing method provided in this application can be executed by a privacy information processing device. This application uses an example of a privacy information processing device executing the privacy information processing method to illustrate the privacy information processing device provided in this application.

[0106] like Figure 8 As shown in the illustration, this application also provides a privacy information processing device 800, which specifically includes: The determination module 801 is used to determine at least one target privacy operation corresponding to each privacy information region based on the privacy information content corresponding to each privacy information region in the image to be processed and the text content associated with the image to be processed. The execution module 802 is used to perform corresponding target privacy operations on each privacy information region in the image to be processed.

[0107] Optionally, when the determining module 801 determines at least one target privacy operation corresponding to each privacy information region based on the privacy information content corresponding to each privacy information region in the image to be processed and the text content associated with the image to be processed, it is specifically used for: Based on the privacy information content corresponding to each privacy information area and the text content associated with the image to be processed, the privacy risk score of the image to be processed, the confidence level of the privacy information corresponding to each privacy information area, and the privacy category are obtained. Based on the privacy risk score and the confidence level of privacy information corresponding to each privacy information area, the target privacy processing strategy corresponding to each privacy information area is determined. Based on the privacy information content corresponding to each privacy information area and the privacy category, at least one target privacy operation is determined from the multiple privacy operations included in the target privacy processing strategy.

[0108] Optionally, when the determining module 801 obtains the privacy risk score of the image to be processed based on the privacy information content corresponding to each privacy information area and the text content associated with the image to be processed, it is specifically used for: Based on the privacy information content corresponding to each privacy information area and the text content associated with the image to be processed, multi-dimensional structured data content is obtained, and the structured data content of each dimension is vectorized and encoded to obtain the feature vector corresponding to the structured data content of each dimension. The feature vectors corresponding to the structured data content of each dimension are input into the privacy risk level assessment model corresponding to each dimension to obtain the first privacy risk level score for each dimension. Based on the first privacy risk level score corresponding to each dimension, determine the mapping relationship between the first privacy risk level score range and the privacy risk level; The mapping relationship and the structured data content of each dimension are input into the privacy risk comprehensive assessment model to obtain the scene label, privacy risk level and second privacy risk level score corresponding to each dimension; Based on the second privacy risk level score and weight coefficient corresponding to each scene label, the privacy risk score of the image to be processed is calculated using a weighted geometric average algorithm.

[0109] Optionally, when the determining module 801 obtains the privacy information confidence level and privacy category corresponding to each privacy information region based on the privacy information content corresponding to each privacy information region and the text content associated with the image to be processed, it is specifically used for: Based on the text content associated with the image to be processed, privacy information is identified in the image to be processed, and multiple privacy information areas and corresponding privacy information content are determined. Based on the privacy information content corresponding to each privacy information area, determine the privacy category corresponding to each privacy information area; The confidence level of privacy information in each privacy information area is calculated using the confidence level calculation method corresponding to the privacy category.

[0110] Optionally, when determining the target privacy processing strategy corresponding to each privacy information region based on the privacy risk score and the privacy information confidence level corresponding to each privacy information region, the determining module 801 is specifically used for: The privacy information region with a confidence level greater than or equal to the confidence threshold is defined as the first privacy information region; Based on the privacy category corresponding to each first privacy information area, the privacy processing strategy corresponding to the privacy category of each first privacy information area is determined from the first privacy processing strategy set; If the privacy risk score is greater than the risk score threshold, a target privacy processing strategy corresponding to each first privacy information region is determined from the first privacy processing strategy set based on the privacy processing strategy corresponding to each first privacy information region. The priority of the target privacy processing strategy is higher than the priority of the first privacy processing strategy. If the privacy risk score is less than or equal to the risk score threshold, a target privacy processing strategy corresponding to each first privacy information region is determined from the first privacy processing strategy set based on the privacy processing strategy corresponding to each first privacy information region. The priority of the target privacy processing strategy is lower than the priority of the first privacy processing strategy.

[0111] Optionally, when determining the target privacy processing strategy corresponding to each privacy information region based on the privacy risk score and the privacy information confidence level corresponding to each privacy information region, the determining module 801 is further configured to: The privacy information region whose confidence level is less than the confidence threshold is defined as the second privacy information region; Based on the mapping relationship between privacy categories and privacy processing strategies in the second privacy processing strategy set, determine the privacy processing strategy corresponding to the privacy category of each second privacy information area from the second privacy processing strategy set; Based on the coordinate information of each second privacy information region, calculate the intersection-union ratio of every two second privacy information regions; Based on the privacy processing strategy corresponding to each second privacy information region and the intersection-union ratio of every two second privacy information regions, the target privacy processing strategy corresponding to each second privacy information region is determined.

[0112] Optionally, when determining the target privacy processing strategy corresponding to each second privacy information region based on the privacy processing strategy corresponding to each second privacy information region and the intersection-union ratio of every two second privacy information regions, the determining module 801 is specifically used for: If the intersection-union ratio of the two second privacy information regions is less than or equal to a preset value, the privacy processing strategy corresponding to each of the two second privacy information regions is determined as the target privacy processing strategy corresponding to each region. If the intersection-union ratio of the two second privacy information regions is greater than the preset value, the conflicting region and the non-conflicting region of the two second privacy information regions are determined. The privacy processing strategy with higher priority among the two privacy processing strategies corresponding to the second privacy information regions is determined as the target privacy processing strategy corresponding to the conflict region. The privacy processing strategies corresponding to the two second privacy information regions are determined as the target privacy processing strategies corresponding to their respective non-conflicting regions.

[0113] Optionally, when determining the target privacy processing strategy corresponding to each second privacy information region based on the privacy processing strategy corresponding to each second privacy information region and the intersection-union ratio of every two second privacy information regions, the determining module 801 is further configured to: If the privacy processing strategies corresponding to the two second privacy information regions have the same priority, the privacy processing strategy corresponding to the privacy operation with higher priority among the privacy operations contained in the privacy processing strategies corresponding to the two second privacy information regions shall be determined as the target privacy processing strategy corresponding to the conflict region.

[0114] Optionally, the device further includes: The processing module is used to reconstruct the privacy processing strategy based on the coordinate information corresponding to each second privacy information region, the privacy category, and the target privacy processing strategy, and add the reconstructed privacy processing strategy to the first privacy processing strategy set.

[0115] Optionally, when the execution module 802 performs corresponding target privacy operations on each privacy information region in the image to be processed, it is specifically used for: Based on the coordinate information corresponding to each privacy information region and the target privacy operation, the operation parameters of the target privacy operation corresponding to each privacy information region are determined. Based on the operation parameters of the target privacy operation corresponding to each privacy information region, the corresponding target privacy operation is performed on each privacy information region in the image to be processed.

[0116] Optionally, the device further includes: The generation module is used to generate interactive structured privacy tags corresponding to each privacy information region based on at least one of the coordinate information corresponding to each privacy information region, the scene tag, the privacy category, and the target privacy processing strategy. The display module is configured to, in response to a triggering operation on a structured privacy label in the privacy information area, display at least one of the coordinate information, the scene label, the privacy category, and the target privacy processing strategy corresponding to the privacy information area.

[0117] The privacy information processing device in this application embodiment can be an electronic device or a component within an electronic device, such as an integrated circuit or a chip. The electronic device can be a terminal or other devices besides a terminal. For example, the electronic device can be a mobile phone, tablet computer, laptop computer, PDA, in-vehicle electronic device, mobile internet device (MID), augmented reality (AR) / virtual reality (VR) device, robot, wearable device, ultra-mobile personal computer (UMPC), netbook, or personal digital assistant (PDA), etc. It can also be a server, network attached storage (NAS), personal computer (PC), television (TV), ATM, or self-service machine, etc. This application embodiment does not specifically limit the device.

[0118] The privacy information processing device in this application embodiment can be a device with an operating system. This operating system can be Android, iOS, or other possible operating systems; this application embodiment does not specifically limit the specific operating system used.

[0119] The privacy information processing device provided in this application embodiment can achieve... Figures 1 to 7 The various processes implemented in the method implementation examples will not be described again here to avoid repetition.

[0120] Optionally, such as Figure 9 As shown, this application embodiment also provides an electronic device 900, including a processor 901 and a memory 902. The memory 902 stores a program or instructions that can run on the processor 901. When the program or instructions are executed by the processor 901, they implement the various steps of the above-described privacy information processing method embodiment and can achieve the same technical effect. To avoid repetition, they will not be described again here.

[0121] It should be noted that the electronic devices in the embodiments of this application include the mobile electronic devices and non-mobile electronic devices described above.

[0122] Figure 10 A schematic diagram of the hardware structure of an electronic device to implement an embodiment of this application.

[0123] The electronic device 1000 includes, but is not limited to, components such as: radio frequency unit 1001, network module 1002, audio output unit 1003, input unit 1004, sensor 1005, display unit 1006, user input unit 1007, interface unit 1008, memory 1009, and processor 1010.

[0124] Those skilled in the art will understand that the electronic device 1000 may also include a power supply (such as a battery) for supplying power to various components. The power supply may be logically connected to the processor 1010 through a power management system, thereby enabling functions such as managing charging, discharging, and power consumption through the power management system. Figure 10 The electronic device structure shown does not constitute a limitation on the electronic device. The electronic device may include more or fewer components than shown, or combine certain components, or have different component arrangements, which will not be elaborated here.

[0125] The processor 1010 is configured to determine at least one target privacy operation corresponding to each privacy information region based on the privacy information content corresponding to each privacy information region in the image to be processed and the text content associated with the image to be processed. Perform corresponding target privacy operations on each privacy information region in the image to be processed.

[0126] Optionally, when the processor 1010 determines at least one target privacy operation corresponding to each privacy information region based on the privacy information content corresponding to each privacy information region in the image to be processed and the text content associated with the image to be processed, it is specifically used for: Based on the privacy information content corresponding to each privacy information area and the text content associated with the image to be processed, the privacy risk score of the image to be processed, the confidence level of the privacy information corresponding to each privacy information area, and the privacy category are obtained. Based on the privacy risk score and the confidence level of privacy information corresponding to each privacy information area, the target privacy processing strategy corresponding to each privacy information area is determined. Based on the privacy information content corresponding to each privacy information area and the privacy category, at least one target privacy operation is determined from the multiple privacy operations included in the target privacy processing strategy.

[0127] Optionally, when the processor 1010 obtains the privacy risk score of the image to be processed based on the privacy information content corresponding to each privacy information area and the text content associated with the image to be processed, it is specifically used for: Based on the privacy information content corresponding to each privacy information area and the text content associated with the image to be processed, multi-dimensional structured data content is obtained, and the structured data content of each dimension is vectorized and encoded to obtain the feature vector corresponding to the structured data content of each dimension. The feature vectors corresponding to the structured data content of each dimension are input into the privacy risk level assessment model corresponding to each dimension to obtain the first privacy risk level score for each dimension. Based on the first privacy risk level score corresponding to each dimension, determine the mapping relationship between the first privacy risk level score range and the privacy risk level; The mapping relationship and the structured data content of each dimension are input into the privacy risk comprehensive assessment model to obtain the scene label, privacy risk level and second privacy risk level score corresponding to each dimension; Based on the second privacy risk level score and weight coefficient corresponding to each scene label, the privacy risk score of the image to be processed is calculated using a weighted geometric average algorithm.

[0128] Optionally, when the processor 1010 obtains the privacy information confidence level and privacy category corresponding to each privacy information region based on the privacy information content corresponding to each privacy information region and the text content associated with the image to be processed, it is specifically used for: Based on the text content associated with the image to be processed, privacy information is identified in the image to be processed, and multiple privacy information areas and corresponding privacy information content are determined. Based on the privacy information content corresponding to each privacy information area, determine the privacy category corresponding to each privacy information area; The confidence level of privacy information in each privacy information area is calculated using the confidence level calculation method corresponding to the privacy category.

[0129] Optionally, when determining the target privacy processing strategy for each privacy information region based on the privacy risk score and the privacy information confidence level corresponding to each privacy information region, the processor 1010 is specifically used for: The privacy information region with a confidence level greater than or equal to the confidence threshold is defined as the first privacy information region; Based on the privacy category corresponding to each first privacy information area, the privacy processing strategy corresponding to the privacy category of each first privacy information area is determined from the first privacy processing strategy set; If the privacy risk score is greater than the risk score threshold, a target privacy processing strategy corresponding to each first privacy information region is determined from the first privacy processing strategy set based on the privacy processing strategy corresponding to each first privacy information region. The priority of the target privacy processing strategy is higher than the priority of the first privacy processing strategy. If the privacy risk score is less than or equal to the risk score threshold, a target privacy processing strategy corresponding to each first privacy information region is determined from the first privacy processing strategy set based on the privacy processing strategy corresponding to each first privacy information region. The priority of the target privacy processing strategy is lower than the priority of the first privacy processing strategy.

[0130] Optionally, when determining the target privacy processing strategy for each privacy information region based on the privacy risk score and the privacy information confidence level corresponding to each privacy information region, the processor 1010 is further configured to: The privacy information region whose confidence level is less than the confidence threshold is defined as the second privacy information region; Based on the mapping relationship between privacy categories and privacy processing strategies in the second privacy processing strategy set, determine the privacy processing strategy corresponding to the privacy category of each second privacy information area from the second privacy processing strategy set; Based on the coordinate information of each second privacy information region, calculate the intersection-union ratio of every two second privacy information regions; Based on the privacy processing strategy corresponding to each second privacy information region and the intersection-union ratio of every two second privacy information regions, the target privacy processing strategy corresponding to each second privacy information region is determined.

[0131] Optionally, when determining the target privacy processing strategy corresponding to each second privacy information region based on the privacy processing strategy corresponding to each second privacy information region and the intersection-union ratio of every two second privacy information regions, the processor 1010 is specifically used for: If the intersection-union ratio of the two second privacy information regions is less than or equal to a preset value, the privacy processing strategy corresponding to each of the two second privacy information regions is determined as the target privacy processing strategy corresponding to each region. If the intersection-union ratio of the two second privacy information regions is greater than the preset value, the conflicting region and the non-conflicting region of the two second privacy information regions are determined. The privacy processing strategy with higher priority among the two privacy processing strategies corresponding to the second privacy information regions is determined as the target privacy processing strategy corresponding to the conflict region. The privacy processing strategies corresponding to the two second privacy information regions are determined as the target privacy processing strategies corresponding to their respective non-conflicting regions.

[0132] Optionally, when determining the target privacy processing strategy corresponding to each second privacy information region based on the privacy processing strategy corresponding to each second privacy information region and the intersection-union ratio of every two second privacy information regions, the processor 1010 is further configured to: If the privacy processing strategies corresponding to the two second privacy information regions have the same priority, the privacy processing strategy corresponding to the privacy operation with higher priority among the privacy operations contained in the privacy processing strategies corresponding to the two second privacy information regions shall be determined as the target privacy processing strategy corresponding to the conflict region.

[0133] Optionally, the processor 1010 is further configured to: Based on the coordinate information corresponding to each second privacy information region, the privacy category, and the target privacy processing strategy, the privacy processing strategy is reconstructed and added to the first privacy processing strategy set.

[0134] Optionally, when the processor 1010 performs corresponding target privacy operations on each privacy information region in the image to be processed, it is specifically used for: Based on the coordinate information corresponding to each privacy information region and the target privacy operation, the operation parameters of the target privacy operation corresponding to each privacy information region are determined. Based on the operation parameters of the target privacy operation corresponding to each privacy information region, the corresponding target privacy operation is performed on each privacy information region in the image to be processed.

[0135] Optionally, after performing corresponding target privacy operations on each privacy information region in the image to be processed, the processor 1010 is further configured to: Based on at least one of the coordinate information corresponding to each privacy information region, the scene label, the privacy category, and the target privacy processing strategy, generate an interactive structured privacy label corresponding to each privacy information region; The display unit 1006 is configured to, in response to a triggering operation of a structured privacy label in the privacy information area, display at least one of the coordinate information, the scene label, the privacy category, and the target privacy processing strategy corresponding to the privacy information area.

[0136] It should be understood that, in this embodiment, the input unit 1004 may include a graphics processing unit (GPU) 10041 and a microphone 10042. The GPU 10041 processes image data of still images or videos obtained by an image capture device (such as a camera) in video capture mode or image capture mode. The display unit 1006 may include a display panel 10061, which may be configured in the form of a liquid crystal display, an organic light-emitting diode, etc. The user input unit 1007 includes a touch panel 10071 and at least one of other input devices 10072. The touch panel 10071 is also called a touch screen. The touch panel 10071 may include a touch detection device and a touch controller. Other input devices 10072 may include, but are not limited to, physical keyboards, function keys (such as volume control buttons, power buttons, etc.), trackballs, mice, joysticks, etc., which will not be described in detail here.

[0137] The memory 1009 can be used to store software programs and various data. The memory 1009 may primarily include a first storage area for storing programs or instructions and a second storage area for storing data. The first storage area may store the operating system, application programs or instructions required for at least one function (such as sound playback, image playback, etc.). Furthermore, the memory 1009 may include volatile memory or non-volatile memory, or both. The non-volatile memory may be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. Volatile memory can be random access memory (RAM), static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM), and direct memory bus RAM (DRRAM). The memory 1009 in this embodiment includes, but is not limited to, these and any other suitable types of memory.

[0138] The processor 1010 may include one or more processing units; optionally, the processor 1010 integrates an application processor and a modem processor, wherein the application processor mainly handles operations involving the operating system, user interface, and applications, and the modem processor mainly handles wireless communication signals, such as a baseband processor. It is understood that the aforementioned modem processor may also not be integrated into the processor 1010.

[0139] This application also provides a readable storage medium storing a program or instructions. When the program or instructions are executed by a processor, they implement the various processes of the above-described privacy information processing method embodiments and achieve the same technical effects. To avoid repetition, they will not be described again here.

[0140] The processor is the processor in the electronic device described in the above embodiments. The readable storage medium includes computer-readable storage media, such as computer read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk.

[0141] This application embodiment also provides a chip, which includes a processor and a communication interface. The communication interface is coupled to the processor. The processor is used to run programs or instructions to implement the various processes of the above-described privacy information processing method embodiments and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0142] It should be understood that the chip mentioned in the embodiments of this application may also be referred to as a system-on-a-chip, system chip, chip system, or system-on-a-chip, etc.

[0143] This application provides a computer program product, which is stored in a storage medium and executed by at least one processor to implement the various processes of the privacy information processing method embodiments described above, and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0144] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element. Furthermore, it should be noted that the scope of the methods and apparatuses in the embodiments of this application is not limited to performing functions in the order shown or discussed, but may also include performing functions substantially simultaneously or in the reverse order, depending on the functions involved. For example, the described methods may be performed in a different order than described, and various steps may be added, omitted, or combined. Additionally, features described with reference to certain examples may be combined in other examples.

[0145] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a computer software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0146] The embodiments of this application have been described above with reference to the accompanying drawings. However, this application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of this application.

Claims

1. A method of processing private information, characterized by, The method comprises the steps of: determining at least one target privacy operation corresponding to each privacy information region in the to-be-processed picture based on the privacy information content corresponding to each privacy information region in the to-be-processed picture and the text content associated with the to-be-processed picture; performing the corresponding target privacy operation on each privacy information region in the to-be-processed picture.

2. The method of claim 1, wherein, The method comprises the steps of: obtaining a privacy risk score of the to-be-processed picture, a privacy information confidence degree corresponding to each privacy information region and a privacy category based on the privacy information content corresponding to each privacy information region and the text content associated with the to-be-processed picture; determining a target privacy processing strategy corresponding to each privacy information region based on the privacy risk score and the privacy information confidence degree corresponding to each privacy information region; determining at least one target privacy operation from a plurality of privacy operations included in the target privacy processing strategy based on the privacy information content corresponding to each privacy information region and the privacy category.

3. The method of claim 2, wherein, The method comprises the steps of: obtaining a plurality of dimensions of structured data content based on the privacy information content corresponding to each privacy information region and the text content associated with the to-be-processed picture, and performing vectorization coding processing on each dimension of structured data content to obtain a feature vector corresponding to each dimension of structured data content; inputting the feature vector corresponding to each dimension of structured data content into a privacy risk level assessment model corresponding to each dimension to obtain a first privacy risk level score corresponding to each dimension; determining a mapping relationship between a first privacy risk level score interval and a privacy risk level based on the first privacy risk level score corresponding to each dimension; inputting the mapping relationship and each dimension of structured data content into a privacy risk comprehensive assessment model to obtain a scene label, a privacy risk level and a second privacy risk level score corresponding to each dimension; calculating the privacy risk score of the to-be-processed picture by using a weighted geometric mean algorithm according to the second privacy risk level score corresponding to each scene label and a weight coefficient.

4. The method of claim 2, wherein, The method comprises the steps of: performing privacy information recognition on the to-be-processed picture based on the text content associated with the to-be-processed picture to determine a plurality of privacy information regions and corresponding privacy information content; determining a privacy category corresponding to each privacy information region based on the privacy information content corresponding to each privacy information region; calculating the privacy information confidence degree of each privacy information region by using a confidence degree calculation method corresponding to the privacy category.

5. The method of claim 2, wherein, The target privacy processing strategy corresponding to each privacy information region is determined based on the privacy risk score and the privacy information confidence corresponding to each privacy information region, and the target privacy processing strategy corresponding to each privacy information region is determined based on the privacy risk score and the privacy information confidence corresponding to each privacy information region. The privacy information region with the privacy information confidence greater than or equal to the confidence threshold is determined as a first privacy information region. The privacy processing strategy corresponding to each first privacy information region is determined from the first privacy processing strategy set based on the privacy category corresponding to each first privacy information region. In a case where the privacy risk score is greater than a risk score threshold, the target privacy processing strategy corresponding to each first privacy information region is determined from the first privacy processing strategy set based on the privacy processing strategy corresponding to each first privacy information region, and the priority of the target privacy processing strategy is higher than the priority of the first privacy processing strategy. In a case where the privacy risk score is less than or equal to the risk score threshold, the target privacy processing strategy corresponding to each first privacy information region is determined from the first privacy processing strategy set based on the privacy processing strategy corresponding to each first privacy information region, and the priority of the target privacy processing strategy is lower than the priority of the first privacy processing strategy.

6. The method of claim 5, wherein, The target privacy processing strategy corresponding to each privacy information region is determined based on the privacy risk score and the privacy information confidence corresponding to each privacy information region, and the target privacy processing strategy corresponding to each privacy information region is determined based on the privacy risk score and the privacy information confidence corresponding to each privacy information region. The privacy information region with the privacy information confidence less than the confidence threshold is determined as a second privacy information region. The privacy processing strategy corresponding to each second privacy information region is determined from the second privacy processing strategy set based on the mapping relationship between the privacy category and the privacy processing strategy in the second privacy processing strategy set. The intersection-over-union of each two second privacy information regions is calculated based on the coordinate information of each second privacy information region. The target privacy processing strategy corresponding to each second privacy information region is determined based on the privacy processing strategy corresponding to each second privacy information region and the intersection-over-union of each two second privacy information regions.

7. The method of claim 6, wherein, The target privacy processing strategy corresponding to each second privacy information region is determined based on the privacy processing strategy corresponding to each second privacy information region and the intersection-over-union of each two second privacy information regions. In a case where the intersection-over-union of the two second privacy information regions is less than or equal to a preset value, the privacy processing strategy corresponding to each of the two second privacy information regions is determined as the target privacy processing strategy corresponding to each of the two second privacy information regions. In a case where the intersection-over-union of the two second privacy information regions is greater than the preset value, the conflict region and the non-conflict region of the two second privacy information regions are determined. The privacy processing strategy with a higher priority among the privacy processing strategies corresponding to the two second privacy information regions is determined as the target privacy processing strategy corresponding to the conflict region. The privacy processing strategy corresponding to each of the two second privacy information regions is determined as the target privacy processing strategy corresponding to each non-conflict region of the two second privacy information regions.

8. The method of claim 7, wherein, The target privacy processing strategy corresponding to each second privacy information region is determined based on the privacy processing strategy corresponding to each second privacy information region and the intersection-over-union of each two second privacy information regions. In the case where the priority of the privacy processing strategies corresponding to the two second privacy information regions is the same, the privacy processing strategy corresponding to the privacy operation with higher priority in the privacy operations included in the privacy processing strategies corresponding to the two second privacy information regions respectively is determined as the target privacy processing strategy corresponding to the conflict region.

9. The method of claim 5, wherein, The method further comprises: The privacy processing strategy is reconstructed based on the coordinate information corresponding to each second privacy information region, the privacy category, and the target privacy processing strategy, and the reconstructed privacy processing strategy is added to the first privacy processing strategy set.

10. The method of claim 1, wherein, The target privacy operation corresponding to each privacy information region in the picture to be processed is performed, comprising: The operation parameter of the target privacy operation corresponding to each privacy information region is determined based on the coordinate information corresponding to each privacy information region and the target privacy operation; The target privacy operation corresponding to each privacy information region in the picture to be processed is performed based on the operation parameter of the target privacy operation corresponding to each privacy information region.

11. The method of claim 3, wherein, After the target privacy operation corresponding to each privacy information region in the picture to be processed is performed, the method further comprises: The interactive structured privacy label corresponding to each privacy information region is generated based on at least one of the coordinate information corresponding to each privacy information region, the scene label, the privacy category, and the target privacy processing strategy; In response to a triggering operation on the structured privacy label in the privacy information region, at least one of the coordinate information corresponding to the privacy information region, the scene label, the privacy category, and the target privacy processing strategy is displayed.

12. An apparatus for processing private information, characterized by comprising: comprises: The determining module is configured to determine at least one target privacy operation corresponding to each privacy information region in a picture to be processed based on the privacy information content corresponding to each privacy information region in the picture to be processed and the text content associated with the picture to be processed. The executing module is configured to perform the target privacy operation corresponding to each privacy information region in the picture to be processed.

13. An electronic device, comprising: The device comprises a processor and a memory, and the memory stores programs or instructions executable on the processor, and the programs or instructions are executed by the processor to implement the steps of the privacy information processing method according to any one of claims 1-11.

14. A readable storage medium, characterized by, The readable storage medium stores programs or instructions, and the programs or instructions are executed by the processor to implement the steps of the privacy information processing method according to any one of claims 1-11.

Images