Artistic composition perspective aided design method and system

Abstract

The invention discloses an artistic composition perspective aided design method and system, relates to the technical field of aided design, and is used for solving the problem that the working efficiency is reduced to a certain extent due to the fact that existing design software still needs a period of waiting time after real-time calculation and element transfer are needed. Obtaining a target image to be subjected to geometric perspective, obtaining related perspective parameters, matching geometric features of the target image with each preset splitting mode feature, and determining a feature matching degree corresponding to each preset splitting mode; according to the feature matching degree corresponding to each preset splitting mode, determining an advanced calculation area during geometric perspective of the target image; through pre-operation, the buffer time during real-time operation is reduced, the creation efficiency is remarkably improved, and the time for repeated modification is reduced. Meanwhile, by recording the operation history and feedback of the user, the system can continuously optimize the algorithm to adapt to the personalized requirements of the user.

Description

Technical Field

[0001] The present invention relates to the field of auxiliary design technology, and more specifically, to an artistic composition perspective auxiliary design method and system. Background Art

[0002] With the rapid development of digital art and design, creative tools and auxiliary software continue to evolve to meet the increasingly complex needs of artists, designers and architects. In this context, the automatic generation of composition and perspective effects has become particularly important, especially when dealing with various types of visual content. Accurate perspective relationships and element layouts are essential to enhance the visual appeal and professionalism of the work.

[0003] Currently, more and more design software is using interactive feedback to help users adjust elements in the composition in real time. The core of this technical background is to enable designers to easily perform various operations such as moving, scaling, deleting or adding elements through a user-friendly interface, and the system can immediately calculate the new perspective effect and display it on the interface. This real-time feedback mechanism greatly accelerates the creative process, allowing designers to get instant visual feedback at every step of the creation.

[0004] The prior art has the following deficiencies:

[0005] Through real-time calculation, existing design software can quickly update the perspective results according to each user operation. For example, when a user moves an element to a new position, the system will immediately recalculate the perspective relationship between the element and other elements. However, due to the need for real-time calculation, after moving the element, it still takes a while to wait, which slows down the work efficiency to a certain extent.

[0006] In view of the above problems, the present invention proposes a solution. Summary of the invention

[0007] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides an artistic composition perspective auxiliary design method and system to solve the problems raised in the above-mentioned background technology.

[0008] To achieve the above object, the present invention provides the following technical solutions:

[0009] An artistic composition perspective auxiliary design method, which obtains a target image to be geometrically perspectived and obtains relevant perspective parameters, matches the geometric features of the target image with the features of each preset splitting mode, and determines the corresponding feature matching degree in each preset splitting mode; the geometric features of the target image are obtained in combination with a plurality of relevant perspective parameters, including a target image vanishing point and a target image viewpoint;

[0010] The pre-calculation area of ​​the target image geometric perspective is determined according to the corresponding feature matching degree in each preset splitting mode.

[0011] In a preferred embodiment, the preset splitting modes include multiple fixed-direction splitting and disordered splitting.

[0012] In a preferred embodiment, the preset splitting pattern feature is a core area of ​​the preset splitting pattern, including a starting area and an ending area corresponding to the preset splitting pattern.

[0013] In a preferred embodiment, a matching method of the geometric features of the target image and the features of each preset splitting pattern is as follows:

[0014] The geometric features of the target image include the target image vanishing point and the target image viewpoint;

[0015] Calculate the relative distance between the target image viewpoint and the centroid of the starting area corresponding to the preset splitting mode;

[0016] And calculate the number of vanishing points of the target image located in the starting area and the ending area of ​​the corresponding preset splitting mode;

[0017] After the relative distance value and the vanishing point quantity value are normalized, the sum is added to determine the feature matching degree between the target image and the corresponding preset splitting mode.

[0018] In a preferred embodiment, determining the pre-calculation area of ​​the target image geometric perspective according to the corresponding feature matching degree in each preset split mode specifically includes the following steps:

[0019] Sort the feature matching degrees corresponding to each preset splitting mode, and select the preset splitting mode with the largest feature matching degree value as the final splitting mode;

[0020] If the final split mode is unordered, no pre-calculation is performed; otherwise, split perspective pre-calculation is performed according to the final split mode.

[0021] In a preferred embodiment, if the feature matching degrees corresponding to multiple preset splitting patterns are greater than a preset matching degree threshold, and the feature matching degree deviations corresponding to multiple preset splitting patterns are less than a preset deviation threshold, the feature matching degrees corresponding to the multiple preset splitting patterns are screened secondary to determine the final splitting pattern.

[0022] In a preferred embodiment, the splitting habit data of the system user is obtained, including the usage ratio of each preset splitting mode of the user, and similar images in the historical database are determined according to the geometric features of the target image, and the usage ratio of each preset splitting mode corresponding to the similar images is obtained. The corresponding weight coefficient of each preset splitting mode is comprehensively determined according to the usage ratio of each preset splitting mode of the user and the usage ratio of each preset splitting mode corresponding to the similar images, and the weight coefficient of the preset splitting mode to be screened twice is extracted. A weighted calculation is performed according to the corresponding feature matching degree, and the preset splitting mode with the largest weighted calculation result is set as the final splitting mode;

[0023] If the final split mode is unordered, no pre-calculation is performed; otherwise, split perspective pre-calculation is performed according to the final split mode.

[0024] An art composition perspective auxiliary design system includes a user interface module, an image processing module, a similarity analysis module, a split mode evaluation module and a data storage module, and the modules are connected by signals;

[0025] The user interface module is used to provide a front-end interface for users to interact with the system and is used for users to input images and related parameters;

[0026] The image processing module is used to preprocess and extract features of the target image, and also includes a Hough transform unit, which is used to implement the Hough transform algorithm and detect straight lines and vanishing points in the image;

[0027] The similarity analysis module is used to compare the target image with the images in the historical database according to the characteristics of the target image to determine similar images;

[0028] The split mode evaluation module is used to evaluate the preset split modes and screen them according to the feature matching degree and user habits. It also includes a weight calculation unit, which is used to calculate the weight coefficient of each preset split mode according to the user's splitting habits and the historical usage ratio of similar images;

[0029] The data storage module is used to store data in the data processing process of the art composition perspective auxiliary design system.

[0030] The technical effects and advantages of the artistic composition perspective auxiliary design method and system of the present invention are as follows:

[0031] While the user is adjusting the elements in the composition in real time, including adding, moving and deleting objects, the system uses edge detection and Hough transform algorithms to quickly identify the vanishing point and center of gravity in the image, and then predicts the user's element adjustment method, loads the perspective effect in advance, and adopts an incremental update strategy. The system only performs perspective calculations on the affected area to ensure fast feedback. Finally, after the user adjusts the elements, it can be displayed in real time on the user interface, allowing designers to see the results of the adjustment immediately.

[0032] The present invention reduces the buffer time during real-time calculation by performing pre-calculation, significantly improves the creation efficiency and reduces the time for repeated revisions. At the same time, by recording the user's operation history and feedback, the system can continuously optimize the algorithm to adapt to the user's personalized needs. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 This is a flow chart of an artistic composition perspective-assisted design method of the present invention;

[0034] Figure 2 The present invention is a structural diagram of an artistic composition perspective auxiliary design system. DETAILED DESCRIPTION

[0035] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0036] While the user is adjusting the elements in the composition in real time, including adding, moving and deleting objects, the present invention uses edge detection and Hough transform algorithms to quickly identify the vanishing point and center of gravity in the image, and then predicts the user's element adjustment method, loads the perspective effect in advance, and adopts an incremental update strategy. The system only performs perspective calculations on the affected area to ensure fast feedback. Finally, after the user adjusts the elements, it can be displayed in real time on the user interface, allowing the designer to see the results of the adjustment immediately, significantly improving the creation efficiency and reducing the time for repeated modifications. At the same time, by recording the user's operation history and feedback, the system can continuously optimize the algorithm to adapt to the user's personalized needs.

[0037] Embodiment 1, as Figure 1 As shown, the present invention provides an art composition perspective auxiliary design method comprising the following steps:

[0038] Obtain the target image to be geometrically perspectived, which can be introduced into the scene or composition manually or through image input, such as a photo or hand-drawn draft, etc., and obtain relevant perspective parameters, such as viewpoint, line of sight, vanishing point and horizontal line, etc. Similarly, perspective parameters can also be manually input or obtained by algorithm calculation. Match the geometric features of the target image with the features of each preset splitting mode to determine the corresponding feature matching degree under each preset splitting mode; the geometric features of the target image are obtained by combining multiple relevant perspective parameters, such as the target image vanishing point and the target image viewpoint, etc.

[0039] The pre-calculation area of ​​the target image geometric perspective is determined according to the corresponding feature matching degree in each preset splitting mode.

[0040] It should be noted that the preset splitting mode can be set according to actual conditions, for example, it may include but is not limited to the following 7 types:

[0041] From top to bottom, from bottom to top, from left to right, from right to left, from inside to outside, from outside to inside, and in no order.

[0042] From top to bottom means that the user's operation starts from the top of the picture and gradually splits or adjusts the elements towards the bottom. It is common in compositions that prioritize the upper area, such as the sky, tree canopy, and building roof.

[0043] Bottom-up means that the user starts from the bottom of the picture and gradually adjusts or splits the elements upwards. It is usually used for adjustments that prioritize the bottom, such as the ground or foreground objects.

[0044] From left to right means that the user starts from the left side of the screen and gradually adjusts the elements to the right.

[0045] From right to left means that the user starts from the right side of the screen and gradually adjusts to the left.

[0046] From inside to outside means that the user first adjusts the core elements in the center of the picture, and then gradually expands outward to adjust the peripheral elements. This method is usually used to emphasize the composition of the main object, such as portraits and the center of buildings.

[0047] From outside to inside means that the user starts adjusting from the edge or periphery and gradually moves towards the center. This is common in large depth of field composition, where the user first processes the large scenery on the periphery and then gradually processes the details in the center.

[0048] Disorder means that the user's operation does not follow a fixed logic, and randomly selects elements in the screen to adjust or delete. This type of user behavior is relatively free and cannot be predicted in a linear way.

[0049] The characteristics of the preset split mode are the core area of ​​the preset split mode, that is, the starting area and the ending area corresponding to the preset split mode. For example, in the top-to-bottom split mode, its characteristics are the starting top area and the ending bottom area, that is, its starting area is the top area, and the ending area is the bottom area; in the outside-to-inside split mode, its characteristics are the starting extension area and the ending internal area, that is, its starting area is the extension area, and the ending area is the internal area, and so on.

[0050] A method for matching the geometric features of the target image with the features of each preset splitting pattern is as follows:

[0051] The geometric features of the target image include the target image vanishing point and the target image viewpoint;

[0052] Calculate the relative distance between the target image viewpoint and the centroid of the starting area corresponding to the preset splitting mode;

[0053] And calculate the number of vanishing points of the target image located in the starting area and the ending area of ​​the corresponding preset splitting mode;

[0054] After the relative distance value and the vanishing point quantity value are normalized, the sum is added to determine the feature matching degree between the target image and the corresponding preset splitting mode.

[0055] It should be noted that the target image viewpoint is the position where the audience's eyes naturally stay, which is the visual center of gravity of the image. If the center of gravity of the image is obvious, the user is likely to start element splitting and adjustment from here. Therefore, this embodiment compares the target image viewpoint with the starting area of ​​the preset splitting mode, thereby reflecting the possibility of starting splitting from the starting area. The smaller the relative distance value, the greater the possibility.

[0056] At the same time, the target image vanishing point will guide the viewer's line of sight. Usually the user will adjust the composition along the direction of the perspective line, and may choose to start from the area where the vanishing point is located ("from inside to outside"), or start from a certain side according to the extension direction of the line (such as "from top to bottom" or "from left to right"). The vanishing point position can be determined by edge detection and Hough transform algorithms, or manually input when the image is acquired.

[0057] It is important to note that the vanishing points on an image depend on the number and direction of parallel lines in the scene, and are determined by groups of parallel lines in different directions in the image. Each group of parallel lines will have a vanishing point, so there can be one or more vanishing points in an image.

[0058] Through Hough transform, the straight lines in the image can be effectively detected and their intersections can be calculated, thereby determining the position of the vanishing point. That is, the edge image is obtained through edge detection, and then the Hough transform is applied to convert it into the parameter space, and finally the intersection of the straight lines is calculated to obtain the vanishing point. This method is a prior art and will not be described in detail here.

[0059] Therefore, this embodiment counts the vanishing points of the target image in the starting area and the ending area of ​​the preset splitting pattern. If there are more target image vanishing points in the starting area and the ending area of ​​the preset splitting pattern, it means that the image is more likely to extend along the perspective direction, and it is more likely to be split according to the preset splitting pattern.

[0060] The pre-calculation area of ​​the target image geometric perspective is determined according to the corresponding feature matching degree in each preset split mode, which specifically includes the following steps:

[0061] Sort the feature matching degrees corresponding to each preset splitting mode, and select the preset splitting mode with the largest feature matching degree value as the final splitting mode;

[0062] If the final split mode is unordered, no pre-calculation is performed; otherwise, split perspective pre-calculation is performed according to the final split mode.

[0063] Example 2. Example 1 of the present invention determines the final splitting mode by matching the geometric features of the target image with the features of each preset splitting mode to determine the feature matching degree corresponding to each preset splitting mode. However, in actual practice, for a certain target image, it is very likely that the feature matching degree values ​​corresponding to multiple preset splitting modes that are finally calculated are similar in size. At this time, there is obviously a certain error in screening only by feature matching degree. Therefore, this example further supplements the above-mentioned problem.

[0064] If the feature matching degrees corresponding to multiple preset splitting modes are greater than the preset matching degree threshold, and the feature matching degree deviations corresponding to multiple preset splitting modes are less than the preset deviation threshold, the feature matching degrees corresponding to the multiple preset splitting modes are screened for the second time to determine the final splitting mode.

[0065] The specific steps include:

[0066] Obtain the splitting habit data of system users, including the usage ratio of each preset splitting mode of the user, and determine the similar images in the historical database according to the geometric features of the target image, obtain the usage ratio of each preset splitting mode corresponding to the similar images, comprehensively determine the corresponding weight coefficient of each preset splitting mode according to the usage ratio of each preset splitting mode of the user and the usage ratio of each preset splitting mode corresponding to the similar images, extract the weight coefficient of the preset splitting mode to be screened for the second time, perform weighted calculation according to the corresponding feature matching degree, and set the preset splitting mode with the largest weighted calculation result as the final splitting mode.

[0067] If the final split mode is unordered, no pre-calculation is performed; otherwise, split perspective pre-calculation is performed according to the final split mode.

[0068] The present invention performs secondary screening of the feature matching degrees of multiple preset splitting patterns, especially when the feature matching degree values ​​are similar, combined with the user's splitting habits and the usage ratio of historical similar images, can significantly reduce the risk of misjudgment, thereby improving the accuracy of the final splitting pattern. By combining feature matching, user habits and historical data, the system forms a comprehensive judgment mechanism that can more intelligently adapt to the splitting needs in different situations and improve the flexibility and practicality of the composition.

[0069] Furthermore, determining similar images in the historical database according to the geometric features of the target image specifically includes the following steps:

[0070] The similarity between the number of vanishing points and the center of gravity of the image in the historical database and the target image in different directions is calculated and compared with the preset similarity threshold, and the images in the historical database that are greater than the preset similarity threshold are marked as similar images.

[0071] It should be noted that the similarity evaluation calculation is a prior art. This embodiment provides a similarity calculation method, including the following steps:

[0072] Perform vanishing point detection, use Hough transform to detect vanishing points, determine the coordinates of the intersection of the straight lines of the target image (i.e., the vanishing points), and retrieve the coordinates of the image vanishing points in the historical database.

[0073] The target image's centroid coordinates are calculated by pixel weighting, and the image's centroid coordinates are retrieved from the historical database.

[0074] Construct the feature vector of each image, including the number of vanishing points, vanishing point coordinates, center of gravity position and other information;

[0075] For the target image and the image in the historical database, the similarity is calculated by Euclidean distance or other distance metrics. The similarity value is between 0 and 1, and the closer to 1, the more similar.

[0076] Based on the calculated similarity values, the image with the highest similarity can be selected for further analysis. If the similarity values ​​of multiple images exceed a preset threshold, they can be considered similar images and comprehensively evaluated using their proportions based on their splitting patterns.

[0077] It should be noted that the detection of vanishing points using Hough transform specifically includes:

[0078] Use the Canny edge detection algorithm to extract the edge and obtain the binary image E;

[0079] Input the edge image E into the Hough transform and calculate the parameters (r, θ) of each edge point (x, y); r = xcos(θ) + ysin(θ); in the parameter space, count the votes of each (r, θ);

[0080] Find the (r, θ) with the most votes in the parameter space, corresponding to the straight line in the image;

[0081] For every two detected straight lines L1 and L2, calculate their intersection (i.e. vanishing point);

[0082] The equation of a straight line can be expressed as: y = -(cos(θ) / sin(θ))x + (r / sin(θ));

[0083] Solving the system of equations gives the vanishing point coordinates (xd, yd).

[0084] The calculation expression of the pixel weighted calculation center of gravity is:

[0085]

[0086] Among them, I(i,j) is the intensity value of the image at pixel (i,j), usually brightness or color value, (Cx, Cy) is the centroid coordinate, H is the height of the image, which indicates the number of pixels in the image in the vertical direction. That is, the number of rows of the image; W is the width of the image, which indicates the number of pixels in the image in the horizontal direction. That is, the number of columns of the image; when calculating the centroid, these two parameters are used to traverse each pixel in the image to calculate the centroid coordinates of the image.

[0087] The calculated vanishing point coordinates, center of gravity coordinates and number of vanishing points are substituted into the feature vector of similarity calculation to determine similar images.

[0088] The corresponding weight coefficient of each preset splitting mode is comprehensively determined according to the usage ratio of each preset splitting mode of the user and the usage ratio of each preset splitting mode corresponding to similar images. The specific method is as follows:

[0089] The usage ratio of each preset splitting mode of the user and the usage ratio of each preset splitting mode corresponding to similar images are marked as a and b respectively, and the importance coefficient calculation formula of each preset splitting mode can be Imp(i)=a+b. In the formula, Imp(i) is the importance coefficient of each preset splitting mode, and i represents the sequence number of each preset splitting mode;

[0090] After calculating the importance coefficients of each preset splitting mode, comparing the importance coefficients of each preset splitting mode;

[0091] Obviously, the greater the usage ratio of each preset splitting mode by users and the usage ratio of each preset splitting mode corresponding to similar images, the more commonly used the preset splitting mode is. At this time, when the matching degree is not much different, the preset splitting mode is more important.

[0092] At the same time, it should be noted that the usage ratio of each preset splitting mode of users and the usage ratio of each preset splitting mode corresponding to similar images can also be weighted and calculated according to actual conditions, which will not be elaborated here.

[0093] The importance coefficients of the preset splitting modes are weighted according to the priority diagram method to determine the weight values ​​of the preset splitting modes.

[0094] For example, if the number of preset splitting modes to be screened twice is 4, the importance coefficients of each preset splitting mode are assigned weights according to the priority diagram method as shown in Table 1 below:

[0095]

[0096]

[0097] Table 1

[0098] After sorting the importance coefficients of the preset splitting modes, they are respectively matched with splitting modes 1, 2, 3, and 4 in order from large to small, so as to obtain the corresponding weight coefficients.

[0099] like Figure 2 As shown, the present invention also discloses an art composition perspective auxiliary design system, which is used to implement the above-mentioned design method, including a user interface module, an image processing module, a similarity analysis module, a split mode evaluation module and a data storage module, and the modules are signal connected;

[0100] The user interface module is used to provide a front-end interface for users to interact with the system, including a toolbar, a canvas area, and a real-time feedback display, helping users input images and related parameters.

[0101] The image processing module is used to preprocess and extract features of the target image for subsequent analysis and calculation. It also includes a Hough transform unit for implementing the Hough transform algorithm for detecting straight lines and vanishing points in the image.

[0102] The similarity analysis module is used to compare the target image with the images in the historical database according to the characteristics of the target image to determine similar images.

[0103] The split mode evaluation module is used to evaluate the preset split modes and screen them according to feature matching and user habits. It also includes a weight calculation unit, which is used to calculate the weight coefficient of each preset split mode according to the user's splitting habits and the historical usage ratio of similar images.

[0104] The data storage module is used to store data in the data processing process of the art composition perspective auxiliary design system.

[0105] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product.

[0106] Those of ordinary skill in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application of the technical solution and the invention constraints. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0107] In addition, each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0108] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0109] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. An artistic composition perspective-assisted design method, characterized in that: The method comprises the following steps: Obtaining a target image to be geometrically perspectived and obtaining relevant perspective parameters, matching the geometric features of the target image with the features of each preset splitting mode, and determining the corresponding feature matching degree under each preset splitting mode; The geometric features of the target image are obtained by combining multiple relevant perspective parameters, including the target image vanishing point and the target image viewpoint; The pre-calculation area of ​​the target image geometric perspective is determined according to the corresponding feature matching degree in each preset splitting mode.

2. The method for assisting artistic composition perspective design according to claim 1, characterized in that: The preset split modes include various fixed-direction splits and random splits.

3. The method for assisting artistic composition perspective design according to claim 1, characterized in that: The preset splitting pattern feature is a core area of ​​the preset splitting pattern, including a starting area and an ending area corresponding to the preset splitting pattern.

4. The method for assisting artistic composition perspective design according to claim 3, characterized in that ； A method for matching the geometric features of the target image with the features of each preset splitting pattern is as follows: The geometric features of the target image include the target image vanishing point and the target image viewpoint; Calculate the relative distance between the target image viewpoint and the centroid of the starting area corresponding to the preset splitting mode; And calculate the number of vanishing points of the target image located in the starting area and the ending area of ​​the corresponding preset splitting mode; After the relative distance value and the vanishing point quantity value are normalized, the sum is added to determine the feature matching degree between the target image and the corresponding preset splitting mode.

5. The method for assisting artistic composition perspective design according to claim 2, characterized in that: The pre-calculation area of ​​the target image geometric perspective is determined according to the corresponding feature matching degree in each preset split mode, which specifically includes the following steps: Sort the feature matching degrees corresponding to each preset splitting mode, and select the preset splitting mode with the largest feature matching degree value as the final splitting mode; If the final split mode is unordered, no pre-calculation is performed; otherwise, split perspective pre-calculation is performed according to the final split mode.

6. The method for assisting artistic composition perspective design according to claim 1, characterized in that: If the feature matching degrees corresponding to multiple preset splitting modes are greater than the preset matching degree threshold, and the feature matching degree deviations corresponding to multiple preset splitting modes are less than the preset deviation threshold, the feature matching degrees corresponding to the multiple preset splitting modes are screened for the second time to determine the final splitting mode.

7. The method for assisting artistic composition perspective design according to claim 6, characterized in that: Obtain the splitting habit data of the system user, including the usage ratio of each preset splitting mode of the user, and determine the similar images in the historical database according to the geometric features of the target image, obtain the usage ratio of each preset splitting mode corresponding to the similar images, comprehensively determine the corresponding weight coefficient of each preset splitting mode according to the usage ratio of each preset splitting mode of the user and the usage ratio of each preset splitting mode corresponding to the similar images, extract the weight coefficient of the preset splitting mode to be screened twice, perform weighted calculation according to the corresponding feature matching degree, and set the preset splitting mode with the largest weighted calculation result as the final splitting mode; If the final split mode is unordered, no pre-calculation is performed; otherwise, split perspective pre-calculation is performed according to the final split mode.

8. An art composition perspective-assisted design system, used to implement an art composition perspective-assisted design method according to any one of claims 1 to 7, characterized in that: It includes a user interface module, an image processing module, a similarity analysis module, a splitting mode evaluation module and a data storage module, and the signal connections between the modules; The user interface module is used to provide a front-end interface for users to interact with the system and is used for users to input images and related parameters; The image processing module is used to preprocess and extract features of the target image, and also includes a Hough transform unit, which is used to implement the Hough transform algorithm and detect straight lines and vanishing points in the image; The similarity analysis module is used to compare the target image with the images in the historical database according to the characteristics of the target image to determine similar images; The split mode evaluation module is used to evaluate the preset split modes and screen them according to the feature matching degree and user habits. It also includes a weight calculation unit, which is used to calculate the weight coefficient of each preset split mode according to the user's splitting habits and the historical usage ratio of similar images; The data storage module is used to store data in the data processing process of the art composition perspective auxiliary design system.

Images