Puzzle image data processing method and apparatus

Abstract

The application provides a puzzle image data processing method and device, the method comprising: obtaining a target image time type corresponding to a target user account of a puzzle game application; obtaining a puzzle image with the image time type as the target image time type from each puzzle image as a current target puzzle image, the image time type of the puzzle image being determined based on image classification of the puzzle image by content segmentation; and adding the target puzzle image into an image recommendation list corresponding to the target user account of the puzzle game application, which can effectively improve the efficiency and reliability of puzzle image processing, and effectively improve the timeliness and pertinence of puzzle image pushing, meeting the personalized needs of users for puzzle completion time preferences and improving the puzzle image processing efficiency and pushing effectiveness.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to a method and apparatus for processing jigsaw puzzle image data. Background Technology

[0002] Jigsaw puzzles are a popular puzzle game where a complete image is broken down into smaller, equally sized pieces to form a complete puzzle. Users must then reassemble these scrambled pieces to recreate the final image. To cater to diverse user needs, jigsaw puzzles are often categorized based on piece size and content. For mobile devices, puzzles are typically divided into smaller pieces of 8x8, 10x10, and 12x12 sizes. Therefore, data processing of these puzzles is necessary to recommend them to users and meet their personalized needs.

[0003] Currently, existing jigsaw puzzle image data processing methods typically classify jigsaw puzzle images based on their content, such as into categories including architecture, food, anime, and animals. Jigsaw puzzle game providers then push content based on users' preferences for image content types.

[0004] However, different users have different preferences for the time it takes to complete a jigsaw puzzle, and existing jigsaw puzzle image data processing methods can only adapt to user type preferences, not time-based preferences, thus failing to meet users' personalized needs. If the average completion time of historical jigsaw puzzle images is calculated and then recommended to users each time the user's time preference is obtained, it will not only affect the efficiency of jigsaw puzzle image processing, but also prevent the push of new jigsaw puzzle images, thereby affecting the efficiency of pushing jigsaw puzzle images to users and reducing user engagement.

[0005] Therefore, there is an urgent need to design a jigsaw puzzle image data processing method that can both meet users' personalized needs for jigsaw puzzle completion time preferences and improve the efficiency of jigsaw puzzle image processing and the effectiveness of push notifications. Summary of the Invention

[0006] In view of this, embodiments of this application provide a method and apparatus for processing jigsaw puzzle image data to eliminate or improve one or more defects existing in the prior art.

[0007] One aspect of this application provides a method for processing jigsaw puzzle image data, including:

[0008] Obtain the target image time and type currently corresponding to the target user account of the jigsaw puzzle game application;

[0009] acquire a jigsaw image of which the image time type is the target image time type from each of the jigsaw images, as a current target jigsaw image, wherein the image time type of the jigsaw image is determined based on image classification of the jigsaw image by a content-based segmentation manner;

[0010] add the target jigsaw image into an image recommendation list corresponding to the target user account of the jigsaw game application.

[0011] In some embodiments of the present application, the target image time type currently corresponding to the target user account of the jigsaw game application is acquired by:

[0012] acquire the jigsaw game average time of the target user account of the jigsaw game application;

[0013] determine the image time type corresponding to the preset time range in which the jigsaw game average time is located from the preset correspondence between each image time type and each preset time range, and determine the image time type as the target image time type currently corresponding to the target user account.

[0014] In some embodiments of the present application, before the target image time type currently corresponding to the target user account of the jigsaw game application is acquired, the method further comprises:

[0015] pre-process the jigsaw image to delete each jigsaw sub-block of the boundary region of the jigsaw image to obtain a corresponding initial image;

[0016] perform image segmentation on the initial image, and perform partition marking on each segmented region of the initial image obtained by the image segmentation;

[0017] if there are different segmented regions belonging to the same content object in each of the segmented regions, merge the different segmented regions belonging to the same content object to form a merged region, and reset a unique identifier for the merged region;

[0018] regard regions with the same current identifier as a set, extract an average feature vector corresponding to each of the sets, and if the similarity between the average feature vectors corresponding to different sets satisfies a similarity threshold, merge the sets with different similarity satisfying the similarity threshold into one set, and reset a unique identifier for the merged set;

[0019] perform feature extraction on the jigsaw image according to each of the current sets to obtain feature data of the jigsaw image;

[0020] determine the image time type to which the jigsaw image belongs according to the feature data of the jigsaw image.

[0021] In some embodiments of the present application, the image segmentation is performed on the initialization image, and each segmented region of the initialization image obtained through the image segmentation is marked with a partition label, including:

[0022] The initialization image is input into a preset image segmentation model, so that the image segmentation model outputs each segmented region corresponding to the initialization image; wherein the image segmentation model includes a SAM model;

[0023] Each segmented region is marked with a partition label, so that each segmented region corresponds to a different unique identifier.

[0024] In some embodiments of the present application, the regions with the same current identifier are taken as a set, and the average feature vector corresponding to each set is extracted. If the similarity between the average feature vectors corresponding to different sets satisfies a similarity threshold, the sets with different similarity satisfying the similarity threshold are merged into one set, and the unique identifier of the merged set is reset, including:

[0025] The regions with the same current identifier are taken as a set, and each image sub-block in each set is input into a preset image feature extraction model in turn, so that the image feature extraction model outputs a multi-dimensional feature vector corresponding to each image sub-block; wherein the image feature extraction model includes a Resnet model and / or a VGG model;

[0026] According to the multi-dimensional feature vector corresponding to each image sub-block in each set, an average feature vector corresponding to each set is determined;

[0027] It is judged whether the similarity between the average feature vectors corresponding to different sets satisfies a similarity threshold. If yes, the sets with different similarity satisfying the similarity threshold are merged into one set, and the unique identifier of the merged set is reset, so that each image sub-block in the same set is marked with the same unique identifier.

[0028] In some embodiments of the present application, the feature extraction is performed on the jigsaw image according to the current sets to obtain the feature data of the jigsaw image, including:

[0029] The number of current sets of the jigsaw image and the number of image sub-blocks corresponding to each set are obtained, wherein the number of image sub-blocks corresponding to the set is the area of the set.

[0030] determining area mean square errors of the first three sets after the sorting; and determining area mean square errors of all the sets of the jigsaw image;

[0031] determining an area total sum proportion of the first three sets according to the number of the sets and the area mean square errors of the first three sets; and determining the number of the sets with areas less than the area threshold, to obtain feature data including the number of the sets, the respective areas of the respective sets, the area mean square errors of the first three sets, the area mean square errors of all the sets, the area total sum proportion of the first three sets, and the number of the sets with areas less than the first area threshold.

[0032] In some embodiments of the present application, determining the image use time type to which the jigsaw image belongs according to the feature data of the jigsaw image comprises:

[0033] determining whether the area of the set with the largest area among the respective sets corresponding to the jigsaw image is equal to or greater than a second area threshold, and if so, determining that the image use time type to which the jigsaw image belongs is a first type; wherein the second area threshold is greater than the first area threshold;

[0034] if the area of the set with the largest area among the respective sets corresponding to the jigsaw image is less than the second area threshold, determining whether the area total sum proportion of the first three sets corresponding to the jigsaw image is equal to or greater than a proportion threshold;

[0035] if the area total sum proportion of the first three sets corresponding to the jigsaw image is equal to or greater than the proportion threshold, determining whether the feature data corresponding to the jigsaw image satisfies a preset condition, wherein the preset condition comprises that the number of the sets corresponding to the jigsaw image is less than a first quantity threshold, the area of the set with the largest area is less than a third area threshold, and the area mean square errors of all the sets are less than a mean square error threshold;

[0036] if the feature data corresponding to the jigsaw image does not satisfy the preset condition, determining that the image use time type to which the jigsaw image belongs is a second type; and if the feature data corresponding to the jigsaw image satisfies the preset condition, determining that the image use time type to which the jigsaw image belongs is a third type;

[0037] If the area proportion of the first three sets corresponding to the jigsaw image is less than the proportion threshold, it is determined whether the number of sets with an area less than a first area threshold corresponding to the jigsaw image exceeds a second number threshold, wherein the second number threshold is greater than the first number threshold; if yes, it is determined that the image use time type to which the jigsaw image belongs is a fourth type.

[0038] The preset use time time ranges corresponding to the first type, the second type, the third type to the fourth type are sequentially decreased.

[0039] Another aspect of the present application provides a jigsaw image data processing device, comprising:

[0040] An image use time type acquisition module is configured to acquire a target image use time type currently corresponding to a target user account of a jigsaw game application.

[0041] A jigsaw image acquisition module is configured to acquire, from each of the jigsaw images, a jigsaw image with an image use time type being the target image use time type, as a current target jigsaw image, wherein the image use time type of the jigsaw image is determined based on image classification of the jigsaw image by a content segmentation manner.

[0042] A jigsaw image recommendation module is configured to add the target jigsaw image into an image recommendation list corresponding to the target user account of the jigsaw game application.

[0043] A third aspect of the present application provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements a jigsaw image data processing method when executing the computer program.

[0044] A fourth aspect of the present application provides a computer readable storage medium, which stores a computer program, wherein the computer program implements a jigsaw image data processing method when executed by a processor.

[0045] A fifth aspect of the present application provides a computer program product, comprising a computer program, wherein the computer program implements the jigsaw image data processing method when executed by a processor.

[0046] The puzzle image data processing method provided in the application can effectively improve the efficiency and reliability of puzzle image processing, and can effectively improve the timeliness and pertinence of puzzle image pushing, which can meet the personalized demand of users for puzzle completion time preference, and improve the puzzle image processing efficiency and pushing effectiveness.

[0047] Additional advantages, objects, and features of the application will be set forth in part in the description which follows, and will become apparent to those skilled in the art upon examination of the following or can be learned by practice of the application. The objects and other advantages of the application can be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.

[0048] It will be understood by those skilled in the art that the objects and advantages of the present application can not be limited to the above and can be more clearly understood from the following detailed description with reference to the accompanying drawings. BRIEF DESCRIPTION OF DRAWINGS

[0049] The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiments of the application and together with the description serve to explain the principles of the application. The components in the drawings are not drawn to scale, but are merely intended to illustrate the principles of the application. To facilitate an understanding of some portions of the application, corresponding portions of the drawings can be exaggerated relative to other portions, namely, parts shown in the drawings can be shown disproportionately large, to more clearly convey the principles of the application. In the drawings:

[0050] Figure 1 The first flowchart of the puzzle image data processing method in an embodiment of the application.

[0051] Figure 2 The second flowchart of the puzzle image data processing method in an embodiment of the application.

[0052] Figure 3 The flowchart of step 020 in the puzzle image data processing method in an embodiment of the application.

[0053] Figure 4 The flowchart of step 040 in the puzzle image data processing method in an embodiment of the application.

[0054] Figure 5 Flowchart of step 050 in the puzzle image data processing method in an embodiment of the present application.

[0055] Figure 6 Flowchart of step 060 in the puzzle image data processing method in an embodiment of the present application.

[0056] Figure 7 Structure diagram of the puzzle image data processing device in an embodiment of the present application.

[0057] Figure 8 Flowchart of the image classification process based on content segmentation in the application example of the present application.

[0058] Fig. 9(a) is the original image in the application example of the present application.

[0059] Fig. 9(b) is the image segmentation result and the numbering of the puzzle sub-blocks corresponding to the original image in Fig. 9(a) in the application example of the present application.

[0060] Figure 10 Fig. 10 is a schematic diagram of the temporary stacking of the puzzle sub-blocks in the puzzle game process in the application example of the present application.

[0061] Fig. 11(a) is a schematic diagram of the segmentation result of the animal target in the application example of the present application.

[0062] Fig. 11(b) is a schematic diagram of the merging result of the target region puzzle sub-block numbering corresponding to the animal target in the application example of the present application.

[0063] Fig. 12(a) is the original image of a butterfly in the application example of the present application.

[0064] Fig. 12(b) is a schematic diagram of the result before numbering merging corresponding to the original image of the butterfly in Fig. 12(a).

[0065] Fig. 12(c) is a schematic diagram of the result after numbering merging corresponding to the original image of the butterfly in Fig. 12(a).

[0066] Figure 13 Fig. 13 is a schematic diagram of the stack formed after feature extraction of Fig. 11(b) in the application example of the present application.

[0067] Fig. 14(a) is the first example of the large monomer type puzzle image provided in the application example of the present application.

[0068] Fig. 14(b) is a schematic diagram of the feature data corresponding to Fig. 14(a).

[0069] Fig. 14(c) is the second example of the large monomer type puzzle image provided in the application example of the present application.

[0070] Fig. 14(d) is a feature data diagram corresponding to Fig. 14(c).

[0071] Fig. 15(a) is a first example of a non-uniform Top3 type of puzzle image provided in the application.

[0072] Fig. 15(b) is a feature data diagram corresponding to Fig. 15(a).

[0073] Fig. 15(c) is a second example of a non-uniform Top3 type of puzzle image provided in the application.

[0074] Fig. 15(d) is a feature data diagram corresponding to Fig. 15(c).

[0075] Fig. 16(a) is a first example of a uniform distribution type of puzzle image provided in the application.

[0076] Fig. 16(b) is a feature data diagram corresponding to Fig. 16(a).

[0077] Fig. 16(c) is a second example of a uniform distribution type of puzzle image provided in the application.

[0078] Fig. 16(d) is a feature data diagram corresponding to Fig. 16(c).

[0079] Fig. 17(a) is a first example of a multi-small-block distribution type of puzzle image provided in the application.

[0080] Fig. 17(b) is a feature data diagram corresponding to Fig. 17(a).

[0081] Fig. 17(c) is a second example of a multi-small-block distribution type of puzzle image provided in the application.

[0082] Fig. 17(d) is a feature data diagram corresponding to Fig. 17(c). DETAILED DESCRIPTION

[0083] In order to make the objects, technical solutions and advantages of the present application clearer, further detailed description will be given to the present application in combination with the embodiments and drawings. Herein, the illustrative embodiments of the present application and the description thereof are used to explain the present application, but not to limit the present application.

[0084] It should be noted that, in order to avoid the present application being obscured by unnecessary details, only the structures and / or processing steps closely related to the solutions according to the present application are shown in the drawings, and other details not closely related to the present application are omitted.

[0085] It should be emphasized that the term "comprises / comprising" when used in this specification is taken to specify the presence of stated features, elements, steps or components but does not preclude the presence or addition of one or more other features, elements, steps, components, or groups thereof.

[0086] It should also be noted that the term "connected" is used herein to include direct connection between two components and indirect connection between two components through an intermediate component.

[0087] Hereinafter, embodiments of the present application will be described with reference to the accompanying drawings. In the drawings, like reference numerals designate like or similar components, or like or similar steps.

[0088] In order to solve the problem that the existing jigsaw image data processing method cannot simultaneously meet the personalized needs of users, ensure the efficiency of jigsaw image processing, and push new jigsaw images, the embodiments of the present application provide a jigsaw image data processing method, a jigsaw image data processing device for executing the jigsaw image data processing method, an electronic device, a computer readable storage medium and a computer program product, which can meet the personalized needs of users for jigsaw completion time preference, improve jigsaw image processing efficiency and push effectiveness.

[0089] The embodiments will be described in detail as follows.

[0090] Based on this, the embodiments of the present application provide a jigsaw image data processing method which can be implemented by a jigsaw image data processing device, as shown in Figure 1 The jigsaw image data processing method specifically includes the following contents:

[0091] Step 100: Obtain a target image time type currently corresponding to a target user account of a jigsaw game application.

[0092] In one implementation of step 100, the user can select the preferred jigsaw game average time in the jigsaw game application, and then send the jigsaw game average time and the user account information corresponding to the user to the jigsaw image data processing device. The jigsaw image data processing device receives the jigsaw game average time and the user account information corresponding to the user to determine the target image time type currently corresponding to the target user account of the jigsaw game application.

[0093] In one or more embodiments of the present application, the user account refers to a unique identifier corresponding to the user in the jigsaw game application. Only one user corresponds to one user account.

[0094] It can be understood that the jigsaw game average time refers to the average time consumed by the user to complete the jigsaw of a single jigsaw image.

[0095] Step 200: obtaining a jigsaw image of which the image time type is the target image time type from each of the jigsaw images as a current target jigsaw image, wherein the image time type of the jigsaw image is determined after the jigsaw image is classified by the content-based segmentation manner.

[0096] In step 200, the jigsaw image data processing device pre-stores a plurality of jigsaw images corresponding to respective image time types in the local image library, and these jigsaw images can be preferably new jigsaw images which have not been selected by the user.

[0097] Correspondingly, when there is a jigsaw image selected as the current target jigsaw image in each of the locally stored jigsaw images, the jigsaw image and the corresponding image time type can be deleted from the local image library, so that the target jigsaw image pushed to the user each time is an image which has not been used by the user.

[0098] Further, in order to avoid the jigsaw images being quickly consumed, when there is a jigsaw image selected as the current target jigsaw image in each of the locally stored jigsaw images, a corresponding relationship can be established between the jigsaw image, the image time type corresponding to the jigsaw image and the target user account in the local storage, so as to mark that these jigsaw images have been pushed to the target user account. Based on this, in a preferred manner of step 200, it can be first determined whether the corresponding relationship between the target user account, the jigsaw image and the image time type corresponding to the jigsaw image is stored in the local storage, if yes, the jigsaw image and the image time type corresponding to the jigsaw image corresponding to the target user account which has been stored in the local storage are excluded from the image library, and then the jigsaw image of which the image time type is the target image time type is obtained from the filtered image library as the current target jigsaw image.

[0099] Step 300: adding the target jigsaw image into the image recommendation list corresponding to the target user account of the jigsaw game application.

[0100] As can be seen from the above description, the jigsaw image data processing method provided by the embodiments of the present application can effectively improve the efficiency and reliability of jigsaw image processing, and can effectively improve the timeliness and pertinence of jigsaw image pushing, which can not only meet the personalized needs of the user for the jigsaw completion time preference, but also improve the jigsaw image processing efficiency and pushing effectiveness.

[0101] In order to further improve the effectiveness and intelligent degree of the target image time type corresponding to the target user account of the jigsaw game application, in a jigsaw image data processing method provided by an embodiment of the present application, referring to Figure 2 , the step 100 of the jigsaw image data processing method specifically includes the following contents:

[0102] Step 110: obtaining a jigsaw game average time corresponding to a target user account of the jigsaw game application.

[0103] Step 120: determining, from a preset correspondence between each image time type and each preset time range, an image time type corresponding to the preset time range in which the jigsaw game average time is located, and determining the image time type as a target image time type currently corresponding to the target user account.

[0104] In order to further improve the effectiveness and reliability of the image classification of the jigsaw image based on the content segmentation method, further improve the efficiency and reliability of the jigsaw image processing, and meet the personalized demand of the user for the jigsaw completion time, in the jigsaw image data processing method provided in the embodiment of the present application, referring to Figure 2 , the step 100 in the jigsaw image data processing method further specifically contains the following content before the step 100:

[0105] Step 010: pre-processing the jigsaw image to delete each jigsaw sub-block in the boundary area of the jigsaw image, and obtaining a corresponding initial image.

[0106] It can be understood that in step 010, in the jigsaw game, the jigsaw sub-blocks located in the four boundary areas of the jigsaw image have distinctive features and are most likely to be completed first. Therefore, in the later calculation, the features of the sub-blocks in the outermost boundary are not included in the statistical category.

[0107] Step 020: image segmentation is performed on the initial image, and each segmentation area of the initial image obtained by the image segmentation is marked.

[0108] Step 030: if there are different segmentation areas belonging to the same content object in each segmentation area, the different segmentation areas belonging to the same content object are merged to form a merged area, and the unique identification of the merged area is reset.

[0109] It can be understood that in step 030, in the process of playing the jigsaw game, the user often temporarily puts the jigsaw sub-blocks belonging to the same target together according to the visual features such as texture and color. Especially when there are animals in the jigsaw image, the user will temporarily put the jigsaw sub-blocks belonging to the same animal together to facilitate subsequent assembly. In order to simulate this user experience, it is necessary to first segment the content objects such as animals in the image, and merge the numbers of the jigsaw sub-blocks where the content objects such as animals are located. If a jigsaw sub-block contains a content object such as an animal, regardless of the size of the area, the number of the jigsaw sub-block is uniformly changed to the number of the jigsaw sub-block where the animal is located.

[0110] Step 040: The same region is identified as a set, and the average feature vector corresponding to each of the sets is extracted. If the similarity between the average feature vectors corresponding to different sets meets the similarity threshold, the sets with similar average feature vectors are merged into one set, and the unique identifier of the merged set is reset.

[0111] Step 050: According to the current set, the feature of the puzzle image is extracted to obtain the feature data of the puzzle image.

[0112] Step 060: According to the feature data of the puzzle image, determine the image type of the puzzle image.

[0113] In order to further improve the reliability and effectiveness of image segmentation on the initialization image and partition marking of each segmentation region of the initialization image obtained by the image segmentation, in the embodiment of the present application, a puzzle image data processing method is provided, which is described as follows: Figure 3 , the step 020 of the puzzle image data processing method specifically includes the following contents:

[0114] Step 021: input the initialization image into the preset image segmentation model, so that the image segmentation model outputs each segmentation region corresponding to the initialization image; wherein the image segmentation model includes: SAM model.

[0115] It can be understood that the SAM model (Segment Anything Model) is an image segmentation large model opened by Meta, which can segment any object in any image without any annotation. It uses a web browser interface to predict masks using user-provided hints (such as punctuation, boxes and text), thereby realizing the segmentation of objects in the image.

[0116] Step 022: Partition marking is performed on each of the segmentation regions, so that each of the segmentation regions corresponds to a different unique identifier.

[0117] Specifically, in the actual puzzle process, due to the position of the puzzle sub-block being disordered, the user often selects the target sub-block according to the integrity of the target in the image and the color, texture and other characteristics of different targets. To this end, a SAM (Segment Anything Model) model is used to segment the image, distinguish different targets in the image, and assign the same number to the puzzle sub-blocks belonging to the same segmentation region. If a puzzle sub-block contains multiple different segmentation regions, the largest segmentation region number is used. The number size has no special meaning, but the puzzle sub-blocks with the same number are considered to be a set that is easily temporarily stacked together during the puzzle game process. Temporarily stacking similar puzzle sub-blocks is a common strategy in puzzle games.

[0118] To further improve the reliability and effectiveness of the image sub-block number merging, in an embodiment of the present application, a puzzle image data processing method is provided, referring to Figure 4 , the step 040 in the puzzle image data processing method specifically includes the following content:

[0119] Step 041: Take the regions with the same identification as a set, and input each image sub-block in each set into a pre-set image feature extraction model in turn, so that the image feature extraction model outputs a multi-dimensional feature vector corresponding to each image sub-block; wherein the image feature extraction model includes a Resnet model and / or a VGG model;

[0120] Step 042: According to the multi-dimensional feature vector corresponding to each image sub-block in each set, determine the average feature vector corresponding to each set.

[0121] Step 043: Determine whether the similarity between the average feature vectors corresponding to different sets meets a similarity threshold, if yes, merge the sets with different similarity meeting the similarity threshold into one set, and reset the unique identification of the merged set, so that each image sub-block in the same set is marked with the same unique identification; if not, do not merge the sets.

[0122] It can be understood that the ResNet model (deep residual network) is a convolutional neural network that solves the gradient vanishing and gradient explosion problems in deep neural network training by introducing residual learning, enabling the network to learn features and representations more deeply and accurately. The VGG (Visual Geometry Group model) model usually includes multiple convolutional layers and fully connected layers, and it can perform well in various image recognition and classification tasks, especially when using pre-trained models for transfer learning.

[0123] That is, in the process of the puzzle game, the user will also temporarily stack image sub-blocks with similar colors and textures together. The application names the set of image sub-blocks with the same number as a stack. In order to simulate the experience in this game process, the features of each image sub-block are calculated first. In actual application, the features of each image sub-block can be extracted by a neural network such as Resnet or VGG, and represented as a multidimensional vector. Then, the average feature vector corresponding to each stack is calculated. Finally, the cosine similarity between the average feature vectors of each two stacks is calculated. If the similarity is greater than a specified threshold, the numbers of the two stacks are merged.

[0124] In order to further improve the reliability and effectiveness of feature extraction of the puzzle image, in an embodiment of the application, a puzzle image data processing method is provided, referring to Figure 5 , the step 050 in the puzzle image data processing method specifically includes the following contents:

[0125] Step 051: obtaining the number of the current sets of the puzzle image and the number of image sub-blocks corresponding to each set, wherein the number of image sub-blocks corresponding to the set is the area of the set;

[0126] Step 052: sorting each set in descending order of area, determining the area mean square deviation of the first three sets after sorting, and determining the area mean square deviation of all the sets of the puzzle image;

[0127] Step 053: determining the area total sum proportion of the first three sets according to the number of the sets and the area mean square deviation of the first three sets, and determining the number of the sets with an area less than an area threshold, to obtain feature data including the number of the sets, the area corresponding to each set, the area mean square deviation of the first three sets, the area mean square deviation of all the sets, the area total sum proportion of the first three sets, and the number of the sets with an area less than the first area threshold.

[0128] In order to further improve the reliability and effectiveness of determining the image use type to which the puzzle image belongs according to the feature data of the puzzle image, in an embodiment of the application, a puzzle image data processing method is provided, referring to Figure 6 , the step 060 in the puzzle image data processing method specifically includes the following contents:

[0129] Step 061: determining whether the area of the largest set in the sets corresponding to the jigsaw image is equal to or greater than a second area threshold value, if yes, performing step 062; if no (i.e. if the area of the largest set in the sets corresponding to the jigsaw image is less than the second area threshold value), performing step 063.

[0130] Step 062: determining that the image use time type to which the jigsaw image belongs is a first type; wherein the second area threshold value is greater than the first area threshold value.

[0131] Step 063: determining whether the area sum proportion of the first three sets corresponding to the jigsaw image is equal to or greater than a proportion threshold value; if yes (if the area sum proportion of the first three sets corresponding to the jigsaw image is equal to or greater than the proportion threshold value), performing step 064; if no (if the area sum proportion of the first three sets corresponding to the jigsaw image is less than the proportion threshold value), performing step 067.

[0132] Step 064: determining whether the feature data corresponding to the jigsaw image meets a preset condition, if no (if the feature data corresponding to the jigsaw image does not meet the preset condition), performing step 065; if yes (if the feature data corresponding to the jigsaw image meets the preset condition), performing step 066.

[0133] The preset condition includes that the number of the sets corresponding to the jigsaw image is less than a first number threshold value, the area of the largest set is less than a third area threshold value, and the area mean square deviation of all the sets is less than a mean square deviation threshold value.

[0134] Step 065: determining that the image use time type to which the jigsaw image belongs is a second type.

[0135] Step 066: determining that the image use time type to which the jigsaw image belongs is a third type.

[0136] Step 067: determining whether the number of the sets corresponding to the jigsaw image and having an area less than a first area threshold value exceeds a second number threshold value, if yes, performing step 068; if no, determining the jigsaw image as a jigsaw image of a useless time type, and storing all the jigsaw images of the useless time type independently.

[0137] The second number threshold value is greater than the first number threshold value.

[0138] Step 068: determining that the image use time type to which the jigsaw image belongs is a fourth type.

[0139] In one or more embodiments of the present application, the preset time ranges corresponding to the first type, the second type, the third type to the fourth type are sequentially decreased.

[0140] The first type can also be referred to as a large monomer type.

[0141] The second type can also be referred to as a non-uniform Top3 type.

[0142] The third type can also be referred to as a uniform Top3 type or a uniform distribution type.

[0143] The fourth type can also be referred to as a multi-small-block distribution type.

[0144] In one or more embodiments of the present application, the first area threshold can be set to between 5% and 10% of the total area of the complete jigsaw image, preferably 6.25%, for example, for an 8*8 jigsaw image, the first area threshold is preferably 4. It can be understood that the area mentioned in the present application refers to the number of division blocks, such as 8*8 division, the area is 64. The specific value can be set according to the actual application situation. The second area threshold can be set to between 40% and 60% of the total area of the complete jigsaw image, preferably 54.69%, for example, for an 8*8 jigsaw image, the second area threshold is preferably 35.

[0145] The third area threshold can be set to between 20% and 30% of the total area of the complete jigsaw image, preferably 31.25%, for example, for an 8*8 jigsaw image, the third area threshold is preferably 20.

[0146] The proportion threshold can be set to between 80% and 100%, preferably 90%.

[0147] The first number threshold can be set to between 20% and 25% of the number of all sets in the complete jigsaw image, preferably 23.44%, for example, for an 8*8 jigsaw image, the second number threshold is preferably 15.

[0148] The second number threshold can be set to between 25% and 30% of the number of all sets in the complete jigsaw image, preferably 26.56%, for example, for an 8*8 jigsaw image, the second number threshold is preferably 17.

[0149] The mean square error threshold can be set as required, for example, for an 8*8 jigsaw image, the mean square error threshold is preferably 5.

[0150] From the software level, the present application also provides a jigsaw image data processing device for executing all or part of the jigsaw image data processing method, which is described with reference toFigure 7 The puzzle image data processing apparatus specifically includes the following contents:

[0151] An image time type obtaining module 10 is configured to obtain a target image time type corresponding to a target user account of a puzzle game application;

[0152] A puzzle image obtaining module 20 is configured to obtain, from each of the puzzle images, a puzzle image with an image time type being the target image time type as a current target puzzle image, wherein the image time type of the puzzle image is determined based on image classification of the puzzle image by a content segmentation manner;

[0153] A puzzle image recommendation module 30 is configured to add the target puzzle image into an image recommendation list corresponding to the target user account of the puzzle game application.

[0154] The embodiments of the puzzle image data processing apparatus provided in the present application can be used to execute the processing procedures of the embodiments of the puzzle image data processing method described above, and the functions thereof will not be repeated here. Please refer to the detailed description of the embodiments of the puzzle image data processing method.

[0155] The part of the puzzle image data processing performed by the puzzle image data processing apparatus can be completed in a server or a client device. Specifically, the processing capacity of the client device and the use scenario of the user can be considered for selection. The present application does not make any limitation in this regard. If all operations are completed in the client device, the client device can further include a processor for specific processing of the puzzle image data processing.

[0156] The client device described above can have a communication module (i.e., a communication unit) and can be communicatively connected with a remote server to realize data transmission with the server. The server can include a server of a task scheduling center side, and can also include a server of an intermediate platform in other implementation scenarios, such as a server of a third-party server platform communicatively connected with the server of the task scheduling center. The server can include a single computer device, or can include a server cluster composed of multiple servers, or a server structure of a distributed device.

[0157] The server and the client device can communicate using any suitable network protocol, including network protocols that have not yet been developed as of the filing date of this application. The network protocol can include, for example, a TCP / IP protocol, a UDP / IP protocol, an HTTP protocol, an HTTPS protocol, and the like. Of course, the network protocol can also include, for example, a RPC protocol (Remote Procedure Call Protocol), a REST protocol (Representational State Transfer), and the like used on top of the above-mentioned protocols.

[0158] As can be seen from the above description, the puzzle image data processing device provided by the embodiments of the present application can effectively improve the efficiency and reliability of puzzle image processing, and can effectively improve the timeliness and pertinence of puzzle image pushing, which can not only meet the personalized needs of users for puzzle completion time preferences, but also improve the puzzle image processing efficiency and pushing effectiveness.

[0159] To further illustrate the embodiments of the above-mentioned puzzle image data processing method, the present application also provides a specific application example of image classification of the puzzle image based on content segmentation corresponding to steps 010 to 060 in the puzzle image data processing method, which is described in detail with reference to Figure 8 The image classification of the puzzle image based on content segmentation specifically includes the following contents:

[0160] Based on the image content, it is divided into common types including buildings, food, animation, animals, etc. Game manufacturers will recommend according to the user's preference for image content. In addition to this conventional operation, in order to maintain the competitiveness of the product, the game manufacturer also needs to identify the preferences of different users based on more detailed user experience, and classify and stratify the users, and recommend different puzzle levels for users with different levels and different preferences. For this purpose, the application example can make more detailed quantitative analysis on the user time preference of the puzzle game, classify the experience of the puzzle game with more fine-grained features, realize more accurate user classification and content pushing, and finally improve the number of daily active users and user stickiness of the game.

[0161] Based on this, in order to solve the problem that the user classification and content pushing of the puzzle game are not fine enough, the application example provides a method for image classification of the puzzle image based on image content understanding and segmentation and statistics:

[0162] 1. Preprocessing (corresponding to step 010)

[0163] In the puzzle game, the puzzle sub-blocks located at the boundary of the image have distinctive features, and are the most likely to be completed first. Therefore, in the later calculation, the features of the sub-blocks in the outermost circle of the boundary are not included in the statistical category.

[0164] In an example, the original image of the puzzle image can be an 8*8 or 10*10 size image.

[0165] 2. Image segmentation and numbering (corresponding to the previous step 020)

[0166] In the actual puzzle process, since the positions of the puzzle sub-blocks are disordered, the user often selects the target sub-blocks according to the integrity of the targets in the image and the color, texture and other features of different targets. Therefore, the SAM (Segment Anything Model) model is used to segment the image, distinguish different targets in the image, and assign the same number to the puzzle sub-blocks belonging to the same segmentation region, as shown in FIG. 9(a) and FIG. 9(b). If a puzzle sub-block contains multiple different segmentation regions, the largest segmentation region is numbered. The number size has no special meaning, but the puzzle sub-blocks with the same number are considered to be easily temporarily stacked together during the puzzle game. Temporarily stacking similar puzzle sub-blocks is a common strategy in the puzzle game. As shown in the two red frame regions of FIG. 9(a) and FIG. 9(b), two sets of sub-blocks that are temporarily stacked together are shown. Figure 10

[0167] 3. Single target number merging (corresponding to the previous step 030)

[0168] During the process of playing the puzzle game, the user often temporarily puts the puzzle sub-blocks belonging to the same target together according to the visual features such as texture and color. Especially when there are animals in the image, the user will temporarily stack the puzzle sub-blocks belonging to the same animal together to facilitate subsequent assembly. In order to simulate this user experience, it is necessary to first segment the animal target (i.e. one kind of content object) in the image, and merge the numbers of the puzzle sub-blocks where the animal target is located to the same number, as shown in FIG. 11(a) and FIG. 11(b).

[0169] The 140 and 131 blocks in FIG. 11(a) are changed to the same number 19 in FIG. 11(b), indicating that the sub-block numbers of the animal target segmentation region are unified. If a puzzle sub-block contains an animal target region, regardless of the size of the region, the number of the puzzle sub-block is uniformly changed to the number of the puzzle sub-block where the animal target is located. As shown in FIG. 11(a), the sub-block in the second row and the first column is numbered 80, which is not the same as the number 140 of the dog head region. However, in FIG. 11(b), the same sub-block in the second row and the first column is numbered 19, which is the same as the number of the dog region.

[0170] ​4. Merge image sub-block numbers (corresponding to step 040 above)

[0171] In jigsaw puzzle games, users often temporarily group image pieces with similar colors and textures together. These sets of image pieces with the same number are called "heaps." To simulate this game experience, the features of each image piece are first calculated separately. In practical applications, features of each image piece can be extracted using neural networks such as ResNet or VGG and represented as multidimensional vectors. Then, the average feature vector corresponding to each heap is calculated. Finally, the cosine similarity between the average feature vectors of any two heaps is calculated. If the similarity is greater than a specified threshold, the numbers of the two heaps are merged. As shown in Figure 12(a), the original butterfly image, Figure 12(b) shows the result before merging the numbers in the original butterfly image of Figure 12(a), and Figure 12(c) shows the result after merging the numbers in the original butterfly image of Figure 12(a). As can be seen from Figure 12(c), after merging, the numbers of the upper and lower butterfly regions are the same.

[0172] 5. Feature extraction (corresponding to step 050 above)

[0173] In one or more embodiments of this application, the set may be referred to as a heap.

[0174] Correspondingly, after the above steps, the entire image is divided into several piles. The feature data to be extracted next includes: the number of piles, the area of ​​each pile (i.e., the total number of image sub-blocks it contains), the mean squared area of ​​the top three piles, the mean squared area of ​​all piles, the percentage of the total area of ​​the top three piles (i.e., the size of the top three piles divided by the total number of image sub-blocks), and the number of piles with an area less than 4, etc. Figure 13 As shown, there are three heaps numbered 19, 24, and 20. The area of ​​heap number 24 is 3, the area of ​​heap number 20 is 13, and the area of ​​heap number 19 is 48. In this example, the total area of ​​the first three heaps accounts for 100%.

[0175] 6. Quantitative classification (corresponding to step 060 above)

[0176] Based on the features extracted above, taking an 8x8 jigsaw puzzle game as an example, the game experience can be divided into the following typical types:

[0177] 1) Large Single Unit Type: The largest pile has an area exceeding 35. That is, the largest pile in terms of area has more than 35 puzzle pieces. For example... Figures 14(a) to 14(d) As shown, this type of large single-unit type may generally take users a long time to complete, and users will usually complete the splicing of large single-unit areas in the second half of the game.

[0178] 2) Top3 type: the sum of the areas of the top three piles accounts for more than 90%, and the area of the first pile is less than 35. According to the different mean square deviations of the image sub-block numbers of the top three piles, this type can be further divided into uniform Top3 and non-uniform Top3 types. Generally, as shown in FIG. 2B, the non-uniform Top3 type of puzzle takes more time than the uniform Top3 type. Figures 15(a) to 15(d)

[0179] 3) Uniform distribution type: the number of piles is less than 15, the area of the first pile is less than 20, and the mean square deviation of the areas of all piles is less than 5. Since the pile area distribution is relatively uniform, and the areas of the piles are not greatly different, as shown in FIG. 2C, this type of puzzle is relatively simple and takes less time. Figures 16(a) to 16(d)

[0180] 4) Multi-small-piece distribution type: the number of piles with an area less than 4 is more than 17. As shown in FIG. 2D, this type of puzzle contains many small pieces with independent textures and colors. If the colors of the piles in the puzzle are different, the game becomes simple. However, if there are many piles with similar colors and textures in the puzzle, the game becomes very difficult. Figures 17(a) to 17(d)

[0181] The embodiments of the present application also provide an electronic device, which can include a processor, a memory, a receiver and a transmitter. The processor is configured to execute the puzzle image data processing method mentioned in the above embodiments. The processor and the memory can be connected through a bus or other means. Taking the bus connection as an example. The receiver can be connected with the processor and the memory through wired or wireless means. It can be understood that, Figures 9(a) to 17(b) the numbers in the above table are only examples, and can be set according to actual application situations.

[0182] The processor can be a central processing unit (CPU). The processor can also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or a combination of the above.

[0183] ​​​The memory, as a non-transitory computer readable storage medium, can be configured to store non-transitory software programs, non-transitory computer executable programs and modules, such as program instructions / modules corresponding to the puzzle image data processing method in the embodiments of the present application. The processor can execute various functions and data processing of the processor by running the non-transitory software programs, instructions and modules stored in the memory, that is, implement the puzzle image data processing method in the above method embodiments.

[0184] The memory can include a program storage area and a data storage area. The program storage area can store an operating system and application programs required by at least one function. The data storage area can store data created by the processor and the like. In addition, the memory can include a high-speed random access memory, and can also include a non-transitory memory, such as at least one magnetic disk storage device, a flash memory device, or other non-transitory solid-state memory device. In some embodiments, the memory can optionally include a memory remotely arranged with respect to the processor, and these remote memories can be connected to the processor through a network. Examples of the above network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0185] One or more modules are stored in the memory, and when executed by the processor, the puzzle image data processing method in the embodiments is executed.

[0186] In some embodiments of the present application, a user equipment can include a processor, a memory and a transceiver unit which can include a receiver and a transmitter, the processor, the memory, the receiver and the transmitter can be connected through a bus system, the memory is configured to store computer instructions, and the processor is configured to execute the computer instructions stored in the memory to control the transceiver unit to transceive signals.

[0187] As an implementation manner, the functions of the receiver and the transmitter in the present application can be implemented by considering a transceiving circuit or a dedicated chip for transceiving, and the processor can be implemented by considering a dedicated processing chip, a processing circuit or a general-purpose chip.

[0188] As another implementation manner, the server provided by the embodiments of the present application can be implemented by considering using a general-purpose computer. That is, program codes for implementing the functions of the processor, the receiver and the transmitter are stored in the memory, and the general-purpose processor implements the functions of the processor, the receiver and the transmitter by executing the codes in the memory.

[0189] The embodiment of the present application further provides a computer readable storage medium, which stores a computer program. The computer program is executed by a processor to implement the steps of the puzzle image data processing method. The computer readable storage medium can be a tangible storage medium, such as a random access memory (RAM), a memory, a read only memory (ROM), an electrically programmable ROM, an electrically erasable programmable ROM, a register, a floppy disk, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.

[0190] The embodiment of the present application further provides a computer program product, which implements the puzzle image data processing method when executed by a processor.

[0191] Those skilled in the art should understand that the exemplary components, systems and methods described in connection with the embodiments disclosed herein can be implemented in hardware, software or a combination thereof. The specific implementation depends on the particular application and design constraints imposed on the overall system. Those skilled in the art can use different methods to implement the described functions for each particular application, but such implementation should not be considered beyond the scope of the present application. When implemented in hardware, it can be, for example, an electronic circuit, an application specific integrated circuit (ASIC), appropriate firmware, a plug-in, a functional card, etc. When implemented in software, the elements of the present application are program or code segments used to perform the required tasks. The program or code segments can be stored in a machine readable medium or transmitted through a data signal carried in a carrier wave in a transmission medium or communication link.

[0192] It should be noted that the present application is not limited to the specific configurations and processes described above and shown in the drawings. For the sake of brevity, the detailed descriptions of known methods are omitted herein. In the above embodiments, several specific steps are described and shown as examples. However, the method processes of the present application are not limited to the specific steps described and shown, and those skilled in the art can make various changes, modifications and additions, or change the order of the steps, after understanding the spirit of the present application.

[0193] In the present application, the features described and / or illustrated for one embodiment can be used in the same way or in a similar way in one or more other embodiments, and / or in combination with or instead of the features of other embodiments.

[0194] The above merely describes the preferred embodiments of the present application, and is not intended to limit the present application. Those skilled in the art can make various changes and modifications to the embodiments of the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method of processing puzzle image data, characterized by, The method comprises: preprocessing a jigsaw image to delete each jigsaw sub-block of a boundary region of the jigsaw image, to obtain a corresponding initial image; performing image segmentation on the initial image, and performing region labeling on each segmented region of the initial image obtained through the image segmentation; if there are different segmented regions belonging to the same content object in each segmented region, merging the different segmented regions belonging to the same content object to form a merged region, and resetting a unique identifier for the merged region; regarding regions with the same current identifier as a set, extracting an average feature vector corresponding to each of the sets, and if the similarity between the average feature vectors corresponding to different sets satisfies a similarity threshold, merging the sets with different similarity satisfying the similarity threshold into one set, and resetting a unique identifier for the merged set; performing feature extraction on the jigsaw image according to the current sets, to obtain feature data of the jigsaw image; determining an image use time type to which the jigsaw image belongs according to the feature data of the jigsaw image; obtaining a target image use time type currently corresponding to a target user account of a jigsaw game application; obtaining a jigsaw image with an image use time type being the target image use time type from each jigsaw image as a current target jigsaw image, wherein the image use time type of the jigsaw image is determined based on image classification of the jigsaw image by a content segmentation method; adding the target jigsaw image into an image recommendation list corresponding to the target user account of the jigsaw game application.

2. The puzzle image data processing method of claim 1, wherein, The method of obtaining a target image use time type currently corresponding to a target user account of a jigsaw game application comprises: obtaining a jigsaw game average use time corresponding to the target user account of the jigsaw game application; from a preset correspondence between each image use time type and each preset use time range, determining an image use time type corresponding to a preset use time range in which the jigsaw game average use time is located, and determining the image use time type as a target image use time type currently corresponding to the target user account.

3. The puzzle image data processing method of claim 1, wherein, The method of performing image segmentation on the initial image, and performing region labeling on each segmented region of the initial image obtained through the image segmentation comprises: inputting the initial image into a preset image segmentation model, so that the image segmentation model outputs each segmented region corresponding to the initial image; wherein the image segmentation model comprises a SAM model; performing region labeling on each segmented region, so that each segmented region corresponds to a different unique identifier.

4. The puzzle image data processing method of claim 1, wherein, The method of regarding regions with the same current identifier as a set, extracting an average feature vector corresponding to each of the sets, and if the similarity between the average feature vectors corresponding to different sets satisfies a similarity threshold, merging the sets with different similarity satisfying the similarity threshold into one set, and resetting a unique identifier for the merged set comprises: The current same-identified regions are taken as a set, and each image sub-block in each of the sets is input into a preset image feature extraction model in sequence, so that the image feature extraction model outputs a multi-dimensional feature vector corresponding to each image sub-block in each of the sets; wherein the image feature extraction model comprises a Resnet model and / or a VGG model; According to the multi-dimensional feature vector corresponding to each image sub-block in each of the sets, an average feature vector corresponding to each of the sets is determined respectively; If the similarity between the average feature vectors corresponding to different sets meets a similarity threshold, the sets whose similarity meets the similarity threshold are merged into one set, and a unique identifier is re-set for the merged set, so that each image sub-block in the same set is marked with the same unique identifier.

5. The puzzle image data processing method of claim 1, wherein, The feature data of the jigsaw image is obtained by extracting features of the jigsaw image according to the current sets, which comprises: The number of the current sets of the jigsaw image and the number of image sub-blocks corresponding to each of the sets are obtained, wherein the number of image sub-blocks corresponding to the set is the area of the set; The sets are sorted in descending order of area, and the area mean square deviation of the first three sets after sorting is determined; and the area mean square deviation of all the current sets of the jigsaw image is determined; According to the number of sets and the area mean square deviation of the first three sets, the area total sum proportion of the first three sets is determined, and the number of sets with an area less than an area threshold is determined, to obtain feature data containing the number of sets, the area corresponding to each of the sets, the area mean square deviation of the first three sets, the area mean square deviation of all the sets, the area total sum proportion of the first three sets, and the number of sets with an area less than the first area threshold.

6. The puzzle image data processing method of claim 5, wherein, The image use type to which the jigsaw image belongs is determined according to the feature data of the jigsaw image, which comprises: If the area of the set with the largest area in the sets corresponding to the jigsaw image is equal to or greater than a second area threshold, the image use type to which the jigsaw image belongs is determined as a first type; wherein the second area threshold is greater than the first area threshold; If the area of the set with the largest area in the sets corresponding to the jigsaw image is less than the second area threshold, it is determined whether the area total sum proportion of the first three sets corresponding to the jigsaw image is equal to or greater than a proportion threshold; If the area total sum proportion of the first three sets corresponding to the jigsaw image is equal to or greater than the proportion threshold, it is determined whether the feature data corresponding to the jigsaw image meets a preset condition, wherein the preset condition comprises that the number of sets corresponding to the jigsaw image is less than a first quantity threshold, the area of the set with the largest area is less than a third area threshold, and the area mean square deviation of all the sets is less than a mean square deviation threshold; If the feature data corresponding to the jigsaw image does not satisfy the preset condition, it is determined that the image use time type to which the jigsaw image belongs is a second type; if the feature data corresponding to the jigsaw image satisfies the preset condition, it is determined that the image use time type to which the jigsaw image belongs is a third type. If the area total proportion of the first three sets corresponding to the jigsaw image is less than the proportion threshold, it is determined whether the number of sets whose area is less than the first area threshold and corresponding to the jigsaw image exceeds a second number threshold, wherein the second number threshold is greater than the first number threshold; if yes, it is determined that the image use time type to which the jigsaw image belongs is a fourth type. The preset use time ranges corresponding to the first type, the second type, the third type to the fourth type are sequentially decreased.

7. A jigsaw image data processing apparatus characterized by comprising: The jigsaw image data processing device is used to perform the following contents: The jigsaw image is preprocessed to delete each jigsaw sub-block in the boundary area of the jigsaw image, and an initialization image corresponding to the jigsaw image is obtained; The initialization image is subjected to image segmentation, and each segmentation region of the initialization image obtained by the image segmentation is subjected to partition marking; If there are different segmentation regions belonging to the same content object in each segmentation region, the different segmentation regions belonging to the same content object are merged to form a merged region, and the merged region is re-set with a unique identifier; Regions with the same current identifier are taken as a set, and an average feature vector corresponding to each set is extracted; if the similarity between the average feature vectors corresponding to different sets satisfies a similarity threshold, the sets with different similarity satisfying the similarity threshold are merged into one set, and the merged set is re-set with a unique identifier; According to the current sets, feature extraction is performed on the jigsaw image to obtain feature data of the jigsaw image; According to the feature data of the jigsaw image, the image use time type to which the jigsaw image belongs is determined; The jigsaw image data processing device further comprises: An image use time type acquisition module is configured to acquire a target image use time type corresponding to a target user account of a jigsaw game application; A jigsaw image acquisition module is configured to acquire, from each jigsaw image, a jigsaw image with an image use time type being the target image use time type, as a current target jigsaw image, wherein the image use time type of the jigsaw image is determined based on image classification of the jigsaw image by a content segmentation manner; A jigsaw image recommendation module is configured to add the target jigsaw image into an image recommendation list corresponding to the target user account of the jigsaw game application.

8. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The processor executes the computer program to implement the jigsaw image data processing method of any one of claims 1 to 6.

9. A computer readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the jigsaw image data processing method of any one of claims 1 to 6.

Images