A file system-based efficient processing method for mass pictures

Abstract

The application discloses a mass picture efficient processing method based on a file system, relates to the technical field of picture efficient processing, and aims to solve the problem that a recommended recovery list cannot accurately reflect the actual use frequency of pictures. After pictures are deleted in batches by a user, the user's browsing time is monitored to evaluate the browsing state, the picture deletion time is collected, and the number of photo albums and the access frequency of the pictures are judged in combination with the browsing state. The access frequency is used to calculate the access activity, the number of photo albums is used to generate the historical access feature, the historical collection state is acquired, the pictures are screened and marked, an evaluation period is set, the number of editing times and the label change frequency of the marked pictures in the period are detected, the editing operation feature is comprehensively analyzed, the marked pictures are sorted to generate a display priority order, the order is output to a recommended recovery interface, and candidate screening and sorting are performed in combination with the dynamic value of the pictures, so that the rationality of the recommended recovery is improved.

Description

Technical Field

[0001] This invention relates to the field of efficient image processing technology, and more specifically, to a method for efficient processing of massive images based on a file system. Background Technology

[0002] In existing image management applications, when users delete images in batches, the system usually only records the deletion operation and lacks further analysis of user behavior. Existing technologies generally can only provide a single "undo deletion" function, or make image recovery recommendations based on time sequence or simple collection marks. However, since users' motivations for deleting images are complex and diverse, recovery methods based solely on deletion time or collection status cannot accurately reflect the user's potential recovery intentions.

[0003] The existing technology has the following shortcomings:

[0004] Currently, existing systems typically do not consider the historical access history, editing operation characteristics, and multi-album storage status of images. This results in the recommended recovery list failing to accurately reflect the actual usage frequency and user attention of images, thereby reducing the rationality and practicality of recovery recommendations and leading to a poor user experience. Therefore, this paper proposes a method for efficient processing of massive images based on the file system.

[0005] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0006] To overcome the aforementioned deficiencies of the prior art, embodiments of the present invention provide a method for efficient processing of massive images based on a file system, which solves the problems mentioned in the background art by using user behavior monitoring and historical access feature analysis.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a method for efficient processing of massive images based on a file system, comprising the following steps: Step S1: When a user deletes a batch of images, monitor the user's browsing time, assess the user's browsing status based on the browsing time, collect the image deletion time, and combine the user's browsing status to determine whether to implement the recommended recovery mechanism. Step S2: When implementing the recommendation recovery mechanism, the number of albums where the image is stored and the access frequency are detected. The access activity of the image is calculated based on the access frequency, and the historical access characteristics of the image are generated by combining the number of albums where the image is stored. Step S3: Obtain the historical collection status of the images, filter and label the images based on historical access characteristics, set an evaluation period, and detect the number of edits and tag change frequency of the labeled images within the evaluation period; Step S4: Analyze the editing operation characteristics of the marked images by combining the number of edits and the frequency of tag changes. Sort the marked images according to the editing operation characteristics and generate the display priority order of the marked images. Output the marked images to the recommended recovery interface based on the display priority order.

[0008] In a preferred embodiment, in step S1, when a user performs a batch deletion operation on images, the time of the batch deletion operation is recorded and stored in the metadata log of the file system, and is recorded as the image deletion time. Collect the total time from the completion of the batch deletion operation to when the user stops browsing, and obtain the user's browsing time; When a user's browsing time is less than or equal to a preset browsing time threshold, the user's browsing status is determined to be short-term browsing. When a user's browsing time exceeds a preset browsing time threshold, the user's browsing status is determined to be prolonged browsing.

[0009] In a preferred embodiment, in step S1, an activation period is set, which is the length of time from the time the user's batch deletion operation occurs that the image is recoverable. Get the current time, and subtract the image deletion time from the current time to get the image deletion cycle; If the image deletion period is less than or equal to the activation period, the image is determined to meet the candidate conditions. If the image deletion period is longer than the activation period, the image is determined not to meet the candidate criteria.

[0010] In a preferred embodiment, in step S1, when the user's browsing state is long-term browsing and the image deletion cycle is less than or equal to the activation cycle, the recommendation recovery mechanism is triggered and images that meet the candidate conditions are included in the candidate set.

[0011] The recommended recovery mechanism is not triggered when a user's browsing status is short-term or the image deletion period is longer than the activation period.

[0012] In a preferred embodiment, in step S2, when performing the recommended recovery mechanism, the image creation or import time is obtained through the file system's metadata log, and the storage time is obtained by subtracting the image creation or import time from the image deletion time. The number of albums containing the images is collected by reading the album management logs of the file system; Collect the access frequency of images in the candidate set. The access frequency is the total number of times an image is accessed by a user within its storage time. The ratio of access frequency to storage time is used to measure the access activity of an image. After standardizing the number of stored albums and access activity, the historical access characteristics of the images are calculated based on the weighted summation method.

[0013] In a preferred embodiment, in step S3, the image's collection attribute field is retrieved through the image management module; If an image has been saved to favorites, the favorites attribute field records the corresponding favorites information; Conversely, the content of the collection attribute field will be empty, and the historical collection status will be set to 0; When the collection attribute field records the corresponding collection information, the number of collection behaviors is counted based on the collection information, and the result of the cumulative count is used as the historical collection status. The product of the standardized historical collection status and the historical access characteristics is used as the image history score. The median of the historical ratings for each image was used as the historical rating threshold.

[0014] In a preferred embodiment, in step S3, if the historical score of the image is greater than the historical score threshold, the image is marked. Conversely, the image is not labeled. Set an evaluation period, use the identifier of the marked image as the search condition, and call up the user's editing behavior of the marked image in the user operation log; Every editing action was recorded during the evaluation period, and the number of editing actions was counted as the number of edits to mark the image. Retrieve tag operation events related to the identifier of the marked image from the user operation log; The ratio of the number of tag operation events to the evaluation cycle is used as the tag change frequency.

[0015] In a preferred embodiment, in step S4, the number of edits and the tag change frequency are standardized to obtain the number of edits factor and the tag change factor, respectively. The editing operation characteristics are calculated using a composite formula based on the number of edits and the tag change factor. The marked images are sorted in descending order based on editing operation characteristics and combined into a display set; The editing difference value is obtained by subtracting the editing operation features of adjacent marked images in the display set, and the maximum value of each editing difference value is used as the priority display benchmark value; Prioritize displaying the adjacent labeled images corresponding to the baseline value, and select the labeled image with the minimum value of the editing operation feature as the layer boundary.

[0016] In a preferred embodiment, in step S4, the editing operation features of the marked image are compared with the editing operation features of the layer boundary, and the marked image is divided into a high priority display layer and a low priority display layer. For high-priority display layer marked images, the order of display priority is determined by the descending sorting of editing operation features, and the marked images are output to the recommended recovery interface in sequence; For marked images in the low-priority display layer, randomly select marked images and output them to the recommended recovery interface.

[0017] The technical effects and advantages of this invention are as follows: This invention monitors user browsing time after a user deletes a batch of images, assesses the user's browsing status based on the browsing time, collects the image deletion time, and combines the user's browsing status to determine whether to check the number of albums where the images are stored and the access frequency. The access frequency is used to calculate the image's access activity, and the number of albums where the images are stored generates historical access characteristics for the images. The historical collection status of the images is obtained, and the images are filtered and marked based on these historical access characteristics. An evaluation period is set, and the number of edits and tag change frequencies of marked images within the evaluation period are detected. The number of edits and tag change frequencies are combined to analyze the editing operation characteristics of the marked images. Based on these editing operation characteristics, the marked images are sorted and a display priority order is generated. Based on the display priority order, the marked images are output to the recommended recovery interface. This achieves accurate identification of the potential recovery intention of the user's deletion behavior, and combines the dynamic value of the images for candidate filtering and sorting, improving the rationality and practicality of the recommended recovery results, and enhancing the user interaction experience and recovery efficiency. Attached Figure Description

[0018] Figure 1 This is a flowchart illustrating the implementation of a method for efficient processing of massive images based on a file system, as described in this invention.

[0019] Figure 2 This is a schematic diagram illustrating the steps of a method for efficient processing of massive images based on a file system according to the present invention. Detailed Implementation

[0020] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] This invention monitors user browsing time after a user deletes a batch of images, assesses the user's browsing status based on the browsing time, collects the image deletion time, and determines whether to check the number of albums containing the images and the access frequency based on the user's browsing status. It calculates the image access activity based on the access frequency, generates historical access features of the images based on the number of albums containing them, obtains the historical collection status of the images, filters and marks the images based on the historical access features, sets an evaluation period, detects the number of edits and tag change frequencies of marked images within the evaluation period, analyzes the editing operation characteristics of marked images based on the combined number of edits and tag change frequencies, sorts the marked images based on the editing operation characteristics, generates a display priority order for marked images, and outputs the marked images to the recommended recovery interface based on the display priority order. This achieves accurate identification of the potential recovery intention of the user's deletion behavior, and improves the rationality and practicality of the recommended recovery results by combining the dynamic value of the images for candidate filtering and sorting.

[0022] Example 1, as Figures 1 to 2 As shown, a method for efficient processing of massive images based on a file system includes the following steps: Step S1: When a user deletes a batch of images, monitor the user's browsing time, assess the user's browsing status based on the browsing time, collect the image deletion time, and combine the user's browsing status to determine whether to implement the recommended recovery mechanism. Step S2: When implementing the recommendation recovery mechanism, the number of albums where the image is stored and the access frequency are detected. The access activity of the image is calculated based on the access frequency, and the historical access characteristics of the image are generated by combining the number of albums where the image is stored. Step S3: Obtain the historical collection status of the images, filter and label the images based on historical access characteristics, set an evaluation period, and detect the number of edits and tag change frequency of the labeled images within the evaluation period; Step S4: Analyze the editing operation characteristics of the marked images by combining the number of edits and the frequency of tag changes. Sort the marked images according to the editing operation characteristics and generate the display priority order of the marked images. Output the marked images to the recommended recovery interface based on the display priority order.

[0023] The specific implementation is as follows:

[0024] In step S1, when a user performs a batch deletion operation on images, the time of the batch deletion operation is recorded and stored in the metadata log of the file system, and is recorded as the image deletion time.

[0025] It should be noted that a file system is a mechanism for organizing, managing, and accessing files and directories stored on a storage medium. It is responsible for maintaining the metadata information of files and providing file reading, writing, deletion, renaming, and access control. In this embodiment, the file system is used to store image data and its related metadata. The metadata log is a structured data collection in the file system used to record file and directory related attributes and operation behaviors, including but not limited to information such as file creation time, import time, modification time, access time, deletion time, size, owner, access permissions, and operation history.

[0026] Collect user browsing time after the batch deletion operation. User browsing time is the total duration from the completion of the batch deletion operation to the user stopping browsing.

[0027] The user's browsing status is evaluated based on the browsing time, which includes short-term browsing and long-term browsing: When a user's browsing time is less than or equal to a preset browsing time threshold, the user's browsing status is determined to be short-term browsing. When a user's browsing time exceeds a preset browsing time threshold, the user's browsing status is determined to be prolonged browsing.

[0028] The browsing time threshold is a preset time parameter used to distinguish between short-term and long-term browsing by users. It is based on statistical analysis of historical user operation data. The average browsing time after a batch deletion operation is calculated and corrected by standard deviation or percentile. For example, when the average browsing time after a user deletion operation is 42 seconds and the standard deviation is 10 seconds, the browsing time threshold is set to 40 seconds as the dividing point between short-term and long-term browsing.

[0029] Set an activation period, which is the length of time from the time the user's batch deletion operation occurs that the images are recoverable. Images deleted within the activation period are included in the candidate set of the recommended recovery mechanism. The activation period is set through statistical analysis of historical user operation data. By analyzing the average recovery time of users for deleted images, the time interval covering the average recovery time is used as the activation period. For example, if the statistics show that the proportion of users who complete the recovery operation within 48 hours after deleting images reaches 95%, the activation period is set to 48 hours to ensure that potential recovery needs are covered, while avoiding an excessively long time that would result in an overly large candidate set.

[0030] Get the current time, and subtract the image deletion time from the current time to get the image deletion cycle; The image deletion period is compared with the activation period. If the image deletion period is less than or equal to the activation period, the image is determined to meet the candidate conditions; if the image deletion period is greater than the activation period, the image is determined not to meet the candidate conditions.

[0031] The system combines user browsing history with the image deletion cycle to determine whether to implement a recommended image recovery mechanism. When a user's browsing status is long-term browsing and the image deletion period is less than or equal to the activation period, the recommendation recovery mechanism is triggered and images that meet the candidate conditions are included in the candidate set. The recommended recovery mechanism is not triggered when a user's browsing status is short-term or the image deletion period is longer than the activation period.

[0032] By collecting user browsing time after a batch deletion operation and combining it with a browsing time threshold to determine the browsing status, and then comparing it with the image deletion cycle within the activation period, it is possible to accurately identify whether the user has a potential intention to restore the image. This avoids redundantly triggering the recommendation mechanism when the user has no need to restore the image, thereby improving system execution efficiency and the relevance of user operations.

[0033] In step S2, when implementing the recommended recovery mechanism, the number of albums containing the images and the frequency of access are checked: The storage time of images in the candidate set is collected. The storage time is the continuous storage time of the image from creation or import to deletion. The image creation or import time is obtained through the metadata log of the file system. The storage time is obtained by subtracting the image creation or import time from the image deletion time. The number of albums where the images are stored is collected by reading the album management logs of the file system. The number of albums where the images are stored is the total number of albums that the images have been added to or associated with during their storage time. It reflects the distribution of images in different categories or organizational structures. The larger the value, the more albums the images are shared or associated with, that is, the greater the attention the images have in the user's history. The access frequency of images in the candidate set is collected. The access frequency is the total number of times an image is accessed by users within its storage time. It reflects the intensity of user use and the degree of interest in the image. The higher the access frequency, the greater the user's attention to the image.

[0034] It should be noted that the album management log is a data set used to record the relationship between pictures and albums and the operation history, including but not limited to information such as when a picture is added to an album, removed from an album, album creation time, album deletion time, and album name. In this embodiment, the album management log is used to obtain the number of albums where the pictures are stored.

[0035] The ratio of access frequency to storage time is used as the access activity of the image; After standardizing the number of albums and access activity, the historical access characteristics of the images are calculated using a weighted summation method. The specific calculation formula is as follows: ; in, Historical visit characteristics To store the number of photo albums, To increase visitor activity, and This is a weighting coefficient used to adjust the relative contribution of access activity and storage time to historical access characteristics. Based on statistical analysis of historical user operation data, it calculates the correlation coefficient between the number of stored albums and access activity in image recovery operations, measuring their contribution to image recovery behavior. The weighting coefficient is assigned according to the contribution results. For example, in the statistical analysis of a certain user group, if the correlation between access activity and recovery operations is 0.7, while the correlation between the number of stored albums is 0.3, then the weighting coefficient can be set to... .

[0036] The higher the access activity or the more albums stored, the larger the historical access characteristic value, indicating that the image has received more attention in the user's historical usage behavior.

[0037] It should be noted that standardization refers to the process of mapping raw data of different physical quantities or different dimensions to a uniform dimension, uniform numerical range or uniform statistical distribution through a specific mathematical transformation. Standardization methods include, but are not limited to, standard linear transformation based on interval scaling, Z-Score standardization based on statistics or normalization method based on nonlinear mapping function. The application methods of standardization will not be elaborated here.

[0038] By detecting the number of albums containing candidate images and their access frequency, and calculating access activity based on access frequency, and then combining standardized processing and weighted summation to generate historical access features, we can comprehensively quantify the degree of attention that images receive in users' historical usage behavior. This provides an objective basis for evaluating the value of candidate images and ensures the rationality of subsequent screening.

[0039] In step S3, the image collection attribute field is retrieved through the image management module. If the image has been collected, the collection attribute field records the corresponding collection information; otherwise, the content of the collection attribute field is empty and the historical collection status is set to 0. When the collection attribute field records the corresponding collection information, the number of collection behaviors is counted based on the collection information, and the result of the cumulative count is used as the historical collection status. The act of saving refers to the user's actions of saving images, such as adding them to a favorites list or other bookmarks.

[0040] It should be explained that the image management module is a functional unit used for unified retrieval, parsing and management of image files. It can retrieve and parse the metadata information of image files, including the image's creation time, modification time, access frequency, and collection status.

[0041] The product of the standardized historical collection status and the historical access characteristics is used as the image history score. The median of the historical ratings for each image is used as the historical rating threshold. Images are then compared with the historical ratings of the images to filter them. If an image's historical rating is greater than the historical rating threshold, the image will be marked. Conversely, the image will not be labeled.

[0042] An evaluation period is set, and the identifier of the marked image is used as the search condition. The user's editing behavior on the marked image is called from the user operation log. The editing behavior refers to the editing operation performed by the user on the marked image. Each editing behavior is recorded within the evaluation period, and the number of editing behaviors is counted as the number of edits to the marked image. For example, editing actions include cropping, rotating, color correction, adding filters, and resizing.

[0043] It should be explained that the evaluation period is a time period used to statistically analyze the characteristics of operations on labeled images. It can be set to a fixed duration or dynamically adjusted according to user operation habits. For example, when users operate frequently, the evaluation period is set to 7 days, and when user operations are sparse, the evaluation period is set to 30 days. The user operation log is a collection of data information used to record the user's operation behavior during the interaction process. It records the user's operation events on the image in a time-sequential manner. In this embodiment, the user operation log is used to obtain the editing behavior and tag operation events of the labeled image within the evaluation period.

[0044] Tag change frequency refers to how frequently a user modifies the tags of a marked image within the evaluation period. Using the tag identifier of the marked image as the search condition, the tag operation events related to the tag identifier of the marked image in the user operation log are retrieved. The ratio of the number of tag operation events to the evaluation cycle is used as the tag change frequency. Tag operation events include adding tags, deleting tags, modifying tag names, and adjusting tag levels or categories.

[0045] By acquiring the historical collection status of images and combining it with historical access characteristics to filter and tag images, and further detecting the number of edits and tag change frequencies of tagged images, a comprehensive judgment on the static and dynamic value of images can be achieved, thereby improving the accuracy of recommendation and recovery results.

[0046] In step S4, the number of edits and the tag change frequency are standardized to obtain the number of edits factor and the tag change factor, respectively. The number of edits factor and the tag change factor are used to calculate the editing operation characteristics using a composite formula: ,in, For the number of edits factor, For label change factor, Features for editing operations; Editing activity characteristics refer to a comprehensive indicator of the level of user activity in interacting with tagged images. A higher editing activity characteristic value indicates that the content of the tagged images is modified and the tags are adjusted more frequently.

[0047] After sorting the marked images in descending order according to the editing operation features, they are combined into a display set. The editing operation features of adjacent marked images in the display set are subtracted to obtain the editing difference value. The maximum value of each editing difference value is used as the priority display benchmark value. Prioritize displaying the labeled images corresponding to the baseline value among adjacent labeled images, and select the labeled image with the minimum value of the editing operation feature as the layer boundary; If the editing operation feature is greater than or equal to the editing operation feature of the layer boundary, the marked image is assigned to a high-priority display layer; otherwise, the marked image is assigned to a low-priority display layer. For high-priority display layers, the images are sorted in descending order of editing operation features as the display priority, and the images are output to the recommended recovery interface in sequence. For marked images in the low-priority display layer, randomly select marked images and output them to the recommended recovery interface.

[0048] The recommended recovery interface is an interactive interface that displays a set of marked images generated based on the filtering and sorting results after a user deletes an image, allowing the user to browse and choose whether to perform a recovery operation.

[0049] Editing operation characteristics are generated by standardizing the number of edits and the frequency of tag changes. Based on sorting and hierarchical strategies, a display priority order is generated to ensure that images that are frequently modified and have their tags adjusted are displayed first. This highlights high-attention images in the recommended recovery interface, improving the rationality of the recovery results and the user's operational efficiency.

[0050] Finally, it should be noted that in this paper, relational terms such as first and second are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations.

[0051] Furthermore, 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 limitation, 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 said element.

[0052] In this document, the singular forms “a,” “an,” and “the” may also include the plural forms unless the context clearly indicates otherwise. It should also be understood that terms such as “comprising / including” or “having” specify the presence of the stated features, integrals, steps, operations, components, parts, or combinations thereof, but do not preclude the possibility of the presence or addition of one or more other features, integrals, steps, operations, components, parts, or combinations thereof. Meanwhile, the term “and / or” as used in this specification includes any and all combinations of the associated listed items.

[0053] The various embodiments in this specification are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. The various embodiments can be combined as needed, and the same or similar parts can be referred to each other.

[0054] The above description of the disclosed embodiments will enable those skilled in the art to make or use various modifications to these embodiments. It will be readily apparent to those skilled in the art that the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for efficient processing of massive images based on a file system, characterized in that: Includes the following steps: Step S1: When a user deletes a batch of images, monitor the user's browsing time, assess the user's browsing status based on the browsing time, collect the image deletion time, and combine the user's browsing status to determine whether to implement the recommended recovery mechanism. Step S2: When implementing the recommendation recovery mechanism, the number of albums where the image is stored and the access frequency are detected. The access activity of the image is calculated based on the access frequency, and the historical access characteristics of the image are generated by combining the number of albums where the image is stored. Step S3: Obtain the historical collection status of the images, filter and label the images based on historical access characteristics, set an evaluation period, and detect the number of edits and tag change frequency of the labeled images within the evaluation period; Step S4: Analyze the editing operation characteristics of the marked images by combining the number of edits and the frequency of tag changes. Sort the marked images according to the editing operation characteristics and generate the display priority order of the marked images. Output the marked images to the recommended recovery interface based on the display priority order.

2. The method for efficient processing of massive images based on a file system according to claim 1, characterized in that: In step S1, when a user performs a batch deletion operation on images, the time of the batch deletion operation is recorded and stored in the metadata log of the file system, and is recorded as the image deletion time. Collect the total time from the completion of the batch deletion operation to when the user stops browsing, and obtain the user's browsing time; When a user's browsing time is less than or equal to a preset browsing time threshold, the user's browsing status is determined to be short-term browsing. When a user's browsing time exceeds a preset browsing time threshold, the user's browsing status is determined to be prolonged browsing.

3. The method for efficient processing of massive images based on a file system according to claim 2, characterized in that: In step S1, an activation period is set, which is the length of time from the time the user's batch deletion operation occurs that the image is recoverable; Get the current time, and subtract the image deletion time from the current time to get the image deletion cycle; If the image deletion period is less than or equal to the activation period, the image is determined to meet the candidate conditions. If the image deletion period is longer than the activation period, the image is determined not to meet the candidate criteria.

4. The method for efficient processing of massive images based on a file system according to claim 3, characterized in that: In step S1, when the user's browsing status is long-term browsing and the image deletion period is less than or equal to the activation period, the recommendation recovery mechanism is triggered and images that meet the candidate conditions are included in the candidate set. The recommended recovery mechanism is not triggered when a user's browsing status is short-term or the image deletion period is longer than the activation period.

5. The method for efficient processing of massive images based on a file system according to claim 1, characterized in that: In step S2, when performing the recommended recovery mechanism, the image creation or import time is obtained through the file system's metadata log, and the storage time is obtained by subtracting the image creation or import time from the image deletion time. The number of albums containing the images is collected by reading the album management logs of the file system; Collect the access frequency of images in the candidate set. The access frequency is the total number of times an image is accessed by a user within its storage time. The ratio of access frequency to storage time is used to measure the access activity of an image. After standardizing the number of stored albums and access activity, the historical access characteristics of the images are calculated based on the weighted summation method.

6. The method for efficient processing of massive images based on a file system according to claim 1, characterized in that: In step S3, the image's collection attribute field is retrieved through the image management module; If an image has been saved to favorites, the favorites attribute field records the corresponding favorites information; Conversely, the content of the collection attribute field will be empty, and the historical collection status will be set to 0; When the collection attribute field records the corresponding collection information, the number of collection behaviors is counted based on the collection information, and the result of the cumulative count is used as the historical collection status. The product of the standardized historical collection status and the historical access characteristics is used as the image history score. The median of the historical ratings for each image was used as the historical rating threshold.

7. The method for efficient processing of massive images based on a file system according to claim 6, characterized in that: In step S3, if the historical score of an image is greater than the historical score threshold, the image is marked. Conversely, the image is not labeled. Set an evaluation period, use the identifier of the marked image as the search condition, and call up the user's editing behavior of the marked image in the user operation log; Every editing action was recorded during the evaluation period, and the number of editing actions was counted as the number of edits to mark the image. Retrieve tag operation events related to the identifier of the marked image from the user operation log; The ratio of the number of tag operation events to the evaluation cycle is used as the tag change frequency.

8. The method for efficient processing of massive images based on a file system according to claim 7, characterized in that: In step S4, the number of edits and the tag change frequency are standardized to obtain the number of edits factor and the tag change factor, respectively. The editing operation characteristics are calculated using a composite formula based on the number of edits and the tag change factor. The marked images are sorted in descending order based on editing operation characteristics and combined into a display set; The editing difference value is obtained by subtracting the editing operation features of adjacent marked images in the display set, and the maximum value of each editing difference value is used as the priority display benchmark value; Prioritize displaying the adjacent labeled images corresponding to the baseline value, and select the labeled image with the minimum value of the editing operation feature as the layer boundary.

9. The method for efficient processing of massive images based on a file system according to claim 8, characterized in that: In step S4, the editing operation features of the marked image are compared with the editing operation features of the layer boundary to divide the marked image into a high-priority display layer and a low-priority display layer. For high-priority display layer marked images, the order of display priority is determined by the descending sorting of editing operation features, and the marked images are output to the recommended recovery interface in sequence; For marked images in the low-priority display layer, randomly select marked images and output them to the recommended recovery interface.

Images