Drawing suggestion generation method and intelligent drawing auxiliary system

Abstract

The invention provides a drawing suggestion generation method and an intelligent drawing auxiliary system, and belongs to the technical field of computer-aided drawing, and the method comprises the steps: carrying out the feature extraction of an obtained drawing image sequence based on a preset image feature extraction model, and obtaining drawing features; determining a drawing style of a current painter, and combining the image features and the drawing style in a preset cue word format to obtain a combined cue word; and inputting the combined cue word into an artificial intelligence generation content model to generate a drawing suggestion. The drawing suggestion is generated based on the drawing style, so that the generated drawing suggestion is matched with the drawing style of the current painter, and the applicability and accuracy of the generated drawing suggestion are improved. Therefore, subsequent painting of the current painter can be guided based on the painting suggestion, and the painting quality and the painting efficiency are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of computer-aided drawing, and in particular to a drawing suggestion generating method and an intelligent drawing assisting system. Background Art

[0002] With the development of computer technology, drawing methods based on computer software have gradually become popular. Although this type of drawing software provides basic auxiliary functions such as basic graphic drawing tools, layer management, simple alignment and distribution functions, most of them only mechanically provide some preset operation options and lack the ability to deeply perceive and intelligently analyze the real-time drawing actions of the draftsman.

[0003] When using this type of software, draftsmen still need to think about how to optimize the drawing details. That is, the quality of the drawing still depends on the draftsmen's professional technical knowledge and artistic aesthetic ability. The software cannot provide the draftsmen with real-time and effective suggestions that fit the drawing intentions, and it is difficult to substantially improve the drawing efficiency and quality.

[0004] Therefore, there is an urgent need to provide a drawing suggestion generation method and an intelligent drawing assistance system that can provide effective drawing suggestions to the draftsman in real time to improve the drawing efficiency and quality. Summary of the invention

[0005] In view of this, it is necessary to provide a drawing suggestion generating method and an intelligent drawing assistance system to solve the technical problem that the prior art cannot provide drawing suggestions to the draftsman in real time, resulting in low drawing quality and efficiency.

[0006] On the one hand, in order to solve the above technical problem, the present invention provides a method for generating a painting suggestion, comprising: Based on a preset image feature extraction model, feature extraction is performed on the acquired painting image sequence to obtain painting features; Determine the painting style of the current painter, and combine the image feature and the painting style in a preset prompt word format to obtain a combined prompt word; The combined prompt words are input into an artificial intelligence generated content model to generate painting suggestions.

[0007] In a possible implementation, the painting feature includes a painting trajectory; the painting image sequence includes multiple frames of painting images; and extracting features from the acquired painting image sequence based on a preset image feature extraction model to obtain the painting feature includes: Determine the lines, line features and brush features in each of the painting images based on a preset image feature extraction model; the line features include line width and line color, and the brush features include brush tilt angle and brush force; Trajectory matching is performed on the lines in the multiple frames of painting images based on the line features and the brush features to determine the painting trajectory.

[0008] In a possible implementation, the painting feature also includes a painting scene and a painting entity, the painting scene includes a figure painting and a landscape painting, when the painting scene is a figure painting, the painting entity is a human body part, when the painting scene is a landscape painting, the painting entity is a landscape entity.

[0009] In a possible implementation, determining the painting style of the current painter includes: Receiving the painting style of the current painter based on a preset style input window; or, Determine the composition features and color matching features of the last frame of the painting image in the painting image sequence, input the composition features and the color matching features into a style recognition model to obtain the painting style; the style recognition model is a model of the mapping relationship between composition feature data, color matching feature data and painting style established based on painting history big data.

[0010] In a possible implementation, the composition feature includes line smoothness data, balance data, and proportion data, and determining the composition feature of the last frame of the painting image in the painting image sequence includes: Performing texture detection on the last frame of the painting image in the painting image sequence by using a gray level co-occurrence matrix to obtain a texture feature map; Based on the Hough change algorithm, straight lines are extracted from the texture feature map to obtain line feature data; Segmenting the texture feature map based on a watershed algorithm to obtain a plurality of texture regions, and matching each of the texture regions with a preset pattern library to obtain pattern feature data; The line feature data is subjected to a smoothness analysis to obtain the line smoothness data, and the pattern feature data is subjected to a position analysis and a proportion analysis to obtain the balance data and the proportion data accordingly.

[0011] In a possible implementation, the color matching feature includes the center value of the main color cluster and the proportion of the main color cluster; determining the color matching feature of the last frame of the painting image in the painting image sequence includes: Obtaining the RGB value of each pixel in the last frame of the painting image in the painting image sequence by a pixel traversal method; Performing color space conversion on the RGB value to obtain the HSV value of each pixel; The HSV values ​​are clustered to obtain the center values ​​of the main color clusters and the proportions of the main color clusters.

[0012] In a possible implementation, the drawing suggestions include suggestions for optimizing the direction of lines, suggestions for creative deformation of graphics, and suggestions for color matching.

[0013] In a possible implementation, the painting feature further includes a brush type; and the method further includes: Determining a theoretical brush type based on the painting scene; When the brush type is different from the theoretical brush type, a brush switching suggestion is generated.

[0014] On the other hand, the present invention also provides an intelligent drawing assistance system, comprising: A camera, used to obtain a sequence of painting images; A host computer includes a memory and a processor, wherein: The memory is used to store programs; The processor is coupled to the memory and is used to execute the program stored in the memory to implement the steps of the method for generating a painting suggestion described in any one of the possible implementations above; A projector is used for projecting and displaying the painting suggestions and the painting image sequences.

[0015] In a possible implementation, the intelligent drawing assistance system further includes a multi-touch film; The multi-touch film is used to operate the painting image sequence in response to the current painter's gestures.

[0016] The beneficial effect of the present invention is that the method for generating painting suggestions provided by the present invention extracts painting features from the acquired painting image sequence, determines the painting style of the current painter, and then combines the painting features and the painting style to obtain a combined prompt word, which includes the painting style of the current painter, so that the painting suggestions generated based on the artificial intelligence generated content model can match the painting style of the current painter, thereby improving the applicability and accuracy of the generated painting suggestions. Therefore, the subsequent painting of the current painter can be guided based on the painting suggestions, thereby improving the painting quality and painting efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0018] Figure 1 A schematic diagram of a flow chart of an embodiment of a method for generating painting suggestions provided by the present invention; Figure 2 For the present invention Figure 1 A schematic flow chart of an embodiment of S101; Figure 3 A schematic diagram of a flow chart of an embodiment of determining composition features provided by the present invention; Figure 4 A schematic diagram of a flow chart of an embodiment of determining color matching features provided by the present invention; Figure 5 A schematic diagram of the structure of an embodiment of the intelligent drawing assistance system provided by the present invention. DETAILED DESCRIPTION

[0019] 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 those skilled in the art without creative work are within the scope of protection of the present invention.

[0020] It should be understood that the schematic drawings are not drawn to scale. The flowchart used in the present invention shows the operations implemented according to some embodiments of the present invention. It should be understood that the operations of the flowchart can be implemented out of order, and the steps without logical context can be reversed in order or implemented simultaneously. In addition, those skilled in the art, under the guidance of the content of the present invention, can add one or more other operations to the flowchart, and can also remove one or more operations from the flowchart. Some of the block diagrams shown in the accompanying drawings are functional entities, which do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor systems and / or microcontroller systems.

[0021] Reference to an "embodiment" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiment may be included in at least one embodiment of the present invention. The appearance of the phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0022] The present invention provides a drawing suggestion generating method and an intelligent drawing assisting system, which are respectively described below.

[0023] Figure 1 A schematic flow chart of an embodiment of the method for generating painting suggestions provided by the present invention is shown in FIG. Figure 1As shown, the painting suggestion generation method includes: S101, extracting features from the acquired painting image sequence based on a preset image feature extraction model to obtain painting features; S102, determining the painting style of the current painter, and combining the image features and the painting style in a preset prompt word format to obtain a combined prompt word; S103, inputting the combined prompt words into an artificial intelligence generated content (AIGC) model to generate a painting suggestion.

[0024] The image feature extraction model in step S101 refers to a model for characterizing the mapping relationship between image and painting features obtained by training an initial feature extraction model using historical image data as samples.

[0025] It should be understood that the model structure of the image feature extraction model in step S102 can be any one of the existing mature structures, such as a convolutional neural network model, an autoencoder, etc.

[0026] It should be noted that painting styles include but are not limited to realistic style, cartoon style, anime style, impressionist style and abstract style.

[0027] The artificial intelligence generated content model in step S103 is a model based on the Transformer architecture.

[0028] It should also be noted that: the painting image sequence includes multiple frames of painting images, and each frame of painting image represents a painting image of the current painter at a certain moment.

[0029] Compared with the prior art, the method for generating painting suggestions provided by the embodiment of the present invention extracts painting features from the acquired painting image sequence, determines the painting style of the current painter, and then combines the painting features and the painting style to obtain a combined prompt word, which includes the painting style of the current painter, so that the painting suggestions generated based on the artificial intelligence generated content model can match the painting style of the current painter, thereby improving the applicability and accuracy of the generated painting suggestions. Therefore, the subsequent painting of the current painter can be guided based on the painting suggestions, thereby improving the painting quality and painting efficiency.

[0030] In some embodiments of the present invention, the painting feature includes a painting trajectory, where the painting trajectory refers to a line formed by combining lines in at least one frame of painting image.

[0031] If Figure 2 As shown, step S101 includes: S201, determining the lines, line features and brush features in each painting image based on a preset image feature extraction model; the line features include line width and line color, and the brush features include brush tilt angle and brush force; S202: performing trajectory matching on lines in multiple frames of painting images based on line features and brush features to determine a painting trajectory.

[0032] The brush force can be determined based on the thickness of the line and the distribution of the ink on the line. In particular, when the brush is a writing brush, the line is thinner when the brush is lifted, and the distribution of the ink on the line is more dispersed.

[0033] The embodiment of the present invention matches the trajectory of lines in multiple frames of painting images to avoid misjudging the same painting trajectory distributed in different frames of painting images as different painting trajectories, thereby ensuring the accuracy of the painting trajectory. In addition, the embodiment of the present invention matches the lines based on the line features and the brush features as matching parameters, further ensuring the accuracy of the painting trajectory.

[0034] Due to different painting types, the paintings are also different. For example, when the painting type is a sketch, the color matching should be a single black. When the painting type is an oil painting, the color matching is a combination of multiple colors.

[0035] Therefore, in a specific embodiment of the present invention, the painting feature also includes the painting type, and the painting type includes sketch, ink painting, oil painting and engineering drawing.

[0036] The embodiment of the present invention can further ensure the accuracy of the generated painting suggestions by identifying the painting type.

[0037] It can be seen from the above description that the accuracy of the painting style is crucial to the accuracy of the generated painting suggestions. For painters with a certain painting foundation, they know their own painting style, while for painters without a painting foundation or painters who are not familiar with the definition of painting style, they cannot directly provide their own painting style. To adapt to the above two different situations, in some embodiments of the present invention, determining the painting style of the current painter in step S102 includes: Receiving the painting style of the current painter based on a preset style input window; or, Determine the composition features and color matching features of the last frame of the painting image in the painting image sequence, input the composition features and color matching features into the style recognition model to obtain the painting style; the style recognition model is a model of the mapping relationship between composition feature data, color matching feature data and painting style established based on the painting history big data.

[0038] The embodiment of the present invention provides two ways to determine the painting style. For painters who know their own painting style, they can directly enter the painting style in the preset style input window to quickly learn the painting style. For painters who do not know their own painting style, the painting style can be automatically learned by identifying the last frame of the painting image in the painting image sequence, so as to ensure that the generated painting suggestion matches the painter, thereby improving the painting quality and painting efficiency.

[0039] Furthermore, the embodiment of the present invention determines the painting style through the features of two dimensions, namely, the composition feature and the color matching feature, thereby further improving the recognition accuracy of the painting style.

[0040] To avoid adverse effects caused by inconsistent input formats, preferably, the preset style input window receives the current painter's painting style in the form of a drop-down menu.

[0041] Different painting styles have different degrees of line fluidity (smoothness), balance and proportion. For example, cartoon-style lines are smoother than realistic-style lines. The proportions of people and animals in realistic style are close to those in the real world. Cartoon-style figures are shorter and rounder, and their composition proportions are different. Realistic-style pictures have white space, so their balance is lower than that of cartoon-style pictures.

[0042] Based on the above characteristics, in some embodiments of the present invention, the composition features include line smoothness data, balance data and proportion data, then Figure 3 As shown, determining the composition features of the last frame of the painting image in the painting image sequence includes: S301, performing texture detection on the last frame of the painting image in the painting image sequence by using a gray level co-occurrence matrix to obtain a texture feature map; S302, extracting straight lines from the texture feature map based on the Hough transform algorithm to obtain line feature data; S303, segmenting the texture feature map based on a watershed algorithm to obtain multiple texture regions, matching each texture region with a preset pattern library to obtain pattern feature data; S304, performing smoothness analysis on the line feature data to obtain line smoothness data, and performing position analysis and proportion analysis on the pattern feature data to obtain balance data and proportion data accordingly.

[0043] The textures in the texture feature map include, but are not limited to, tree trunk and branch textures, mountain rock textures, and river ripple textures.

[0044] Line feature data includes but is not limited to mountain boundary lines, river boundary lines, etc.

[0045] Texture areas include but are not limited to mountain areas, tree areas, rock areas, and mission areas.

[0046] In a specific embodiment of the present invention, the smoothness analysis in step S304 is specifically as follows: performing Fourier transform on the line feature data, converting the line feature data from the spatial domain to the frequency domain to obtain frequency components, obtaining an amplitude spectrum based on the frequency components, selecting the frequency component before the cutoff of the frequency component index as the low-frequency component according to a preset frequency component cutoff ratio, selecting the frequency component after the cutoff of the frequency component index as the high-frequency component, taking the ratio of the low-frequency component to the high-frequency component as the line smoothness, and taking the average of the line smoothness of all lines as the line smoothing data.

[0047] In a specific embodiment of the present invention, the position analysis in step S304 is specifically as follows: determining the pattern geometric center of each pattern in the pattern feature data, and calculating the center distance between the geometric centers of each pair of patterns, comparing the pattern distance standard deviation with the pattern distance average value, when the difference between the pattern distance standard deviation and the pattern distance average value is greater than a preset difference threshold, the composition distribution is a pattern visually prominent, when the difference between the pattern distance standard deviation and the pattern distance average value is less than or equal to the preset difference threshold, the composition distribution is a pattern uniformly distributed. The image of the work is divided by the method of thirds to obtain the coordinates of the intersection of the thirds line, when the pattern is visually prominent, the pattern with the largest distance between adjacent patterns is selected as the visually prominent pattern, and the degree of conformity between the visually prominent pattern and the intersection of the thirds line is calculated as the composition balance data, and when the pattern is uniformly distributed, the mean degree of conformity between the geometric center of each pattern and the intersection of the thirds line is calculated as the composition balance data.

[0048] Among them, the compliance is:

[0049] In the formula, d c is the distance between the geometric center of the pattern and the coordinates of the intersection of the three-point line; d m Half the length of the canvas diagonal.

[0050] In a specific embodiment of the present invention, the proportion analysis in step S304 is specifically as follows: sorting the pattern size according to the number of pixels in the pattern feature data, obtaining the first preset pattern as the main pattern according to the preset number of main patterns, subtracting the horizontal coordinates of the leftmost and rightmost pixels of the main pattern to obtain the main pattern width, and subtracting the vertical coordinates of the uppermost and lowermost pixels of the main pattern to obtain the main pattern height. The pattern width ratio and pattern height ratio are obtained according to the main pattern width and the main pattern height. That is, the proportion data includes the pattern width ratio and the pattern height ratio.

[0051] In a specific embodiment of the present invention, the color matching features include the center value of the main color cluster and the proportion of the main color cluster; Figure 4 As shown, determining the color matching features of the last frame of the painting image in the painting image sequence includes: S401, obtaining the RGB value of each pixel in the last frame of the painting image in the painting image sequence by a pixel traversal method; S402, performing color space conversion on the RGB value to obtain the HSV value of each pixel; S403, clustering the HSV values ​​to obtain the center value of the main color cluster and the proportion of the main color cluster.

[0052] In a specific embodiment of the present invention, when the painting style is realistic, the theoretical center values ​​of the main color clusters and the proportions of the main color clusters are: color cluster 1, i.e. earth tones: H: 20-40, S: 50-70, V: 60-80, accounting for 30%-50%, color cluster 2, i.e. natural green tones: H: 80-100, S: 60-80, V: 40-60, accounting for 20%-40%, color cluster 3, i.e. sky blue tones: H: 200-220, S: 40-60, V: 70-90, accounting for 10%-30%, color cluster 4, highlight tones: H: 0-20, S: 0-20, V: 90-100, accounting for 5%-15%, color cluster 5, shadow tones: H: 0-20, S: 0-20, V: 10-30, accounting for 5%-15%.

[0053] That is, different painting styles have different color matching characteristics, so the painting style can be identified based on the color matching characteristics.

[0054] In a specific embodiment of the present invention, the drawing suggestions include suggestions for optimizing the direction of lines, suggestions for creative deformation of graphics, and suggestions for color matching.

[0055] Specifically, when the draftsman is drawing a landscape painting and prefers a realistic style, the subsequent drawing method that best fits the realistic style and can enhance the overall aesthetic of the picture can be analyzed, such as suggesting to adjust the layering of the mountains, express the light and shadow effects through a specific line arrangement, or recommend color combinations that match the current natural scenery to fill the sky, grass and other areas. For another example, when drawing an abstract painting, according to the creative rules of abstract art and the emotions or themes that the draftsman wants to convey, some unique graphic deformation suggestions will be generated to make the picture more artistic and personalized. For another example, when drawing a fantasy-style painting, according to the exaggerated shapes, unique color symbols and mysterious atmosphere creation common in fantasy styles, it is judged that the lines of the currently outlined creature in certain parts (such as wings, tails, etc.) can be more exaggerated to enhance the fantasy, so the corresponding painting suggestions are generated, such as "It is recommended to draw the edge lines of the wings more wavy to show the smart flying posture and make the whole creature look more fantastic."

[0056] Different painting types require different brushes to ensure the quality of the drawing. For example, oil paintings need to be drawn with oil brushes, sketches need to be drawn with pencils, and different types of pencils need to be changed during the sketching process.

[0057] To make the generated painting higher in quality, in some embodiments of the present invention, the painting feature further includes a brush type; the painting suggestion generating method further includes: Determine theoretical brush type based on painting scene; Generates brush switching suggestions when the brush type and theoretical brush type are not the same.

[0058] The embodiment of the present invention generates a brush switching suggestion by setting a setting when the brush type is not a theoretical brush type, thereby ensuring that a suitable brush is used for subsequent drawing, thereby further improving the drawing quality.

[0059] In summary, the drawing suggestion generation method provided by the embodiment of the present invention can deeply understand the intention of the draftsman and generate highly creative and professional drawing suggestions based on different drawing styles, which greatly improves the intelligent level of drawing, reduces the time cost of the draftsman in the conception and decision-making process, and improves drawing efficiency.

[0060] On the other hand, the embodiment of the present invention also provides an intelligent drawing assistance system, such as Figure 5 As shown, the intelligent drawing assistance system 500 includes: Camera 501, used to obtain a painting image sequence; The host 502 includes a memory 5021 and a processor 5022, wherein: A memory 5021, used for storing programs; A processor 5022, coupled to the memory 5021, configured to execute a program stored in the memory 5021 to implement the method for generating a painting suggestion in any one of the above embodiments; The projector 503 is used to project and display the painting suggestions and the painting image sequence.

[0061] The embodiment of the present invention projects the drawing suggestions and the drawing image sequence in real time by setting up the projector 503, so that the draftsman can see the drawing suggestions at a glance and intuitively see the drawing suggestions while drawing, without interrupting the drawing operation to obtain the suggestions, thus achieving real-time interaction during the drawing process and further improving the drawing efficiency and quality.

[0062] The communication mode between the host 502 and the camera 501 and the projector 503 can be wired communication or wireless communication. Specifically, the wired communication can be a USB cable or an HDMI cable connection. The wireless communication can be a Wifi connection or a Bluetooth connection.

[0063] Since it is necessary to perform operations such as translation, rotation, and scaling on the painting image sequence during the drawing process, in some embodiments of the present invention, Figure 5 As shown, the intelligent drawing assistance system 500 further includes a multi-touch film 504; The multi-touch film 504 is used to operate the painting image sequence in response to the current painter's gestures.

[0064] The embodiment of the present invention provides a multi-touch film 504, so that the draftsman can perform boundary operations on the drawing content at any time according to his own observations and ideas, making the drawing process an interactive and flexible creative experience, further stimulating the draftsman's creative inspiration, making drawing no longer a one-way drawing behavior, but a creative process full of interaction and exploration. It helps the draftsman to create more creative and higher-quality drawings, and also makes the drawing process more smooth and interactive.

[0065] Those skilled in the art will appreciate that all or part of the processes of the above-mentioned embodiments can be implemented by instructing related hardware (such as a processor, a controller, etc.) through a computer program, and the computer program can be stored in a computer-readable storage medium, wherein the computer-readable storage medium is a disk, an optical disk, a read-only storage memory, or a random access memory, etc.

[0066] The above is a detailed introduction to a method for generating painting suggestions and an intelligent drawing assistance system provided by the present invention. Specific examples are used herein to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scopes, and the content of this specification should not be understood as limiting the present invention.

Claims

1. A method for generating painting suggestions, characterized in that: include: Based on a preset image feature extraction model, feature extraction is performed on the acquired painting image sequence to obtain painting features; Determine the painting style of the current painter, and combine the image feature and the painting style in a preset prompt word format to obtain a combined prompt word; The combined prompt words are input into an artificial intelligence generated content model to generate painting suggestions.

2. The method for generating a painting suggestion according to claim 1, characterized in that: The painting feature includes a painting trajectory; the painting image sequence includes multiple frames of painting images; the feature extraction of the acquired painting image sequence based on a preset image feature extraction model to obtain the painting feature includes: Determine the lines, line features and brush features in each of the painting images based on a preset image feature extraction model; the line features include line width and line color, and the brush features include brush tilt angle and brush force; Trajectory matching is performed on the lines in the multiple frames of painting images based on the line features and the brush features to determine the painting trajectory.

3. The method for generating a painting suggestion according to claim 2, characterized in that: The painting features also include painting types, and the painting types include sketches, ink paintings, oil paintings, and engineering drawings.

4. The method for generating a painting suggestion according to claim 1, characterized in that: Determining the painting style of the current painter includes: Receiving the painting style of the current painter based on a preset style input window; or, Determine the composition features and color matching features of the last frame of the painting image in the painting image sequence, input the composition features and the color matching features into a style recognition model to obtain the painting style; the style recognition model is a model of the mapping relationship between composition feature data, color matching feature data and painting style established based on painting history big data.

5. The method for generating a painting suggestion according to claim 4, characterized in that: The composition features include line smoothness data, balance data and proportion data, and determining the composition features of the last frame of the painting image in the painting image sequence includes: Performing texture detection on the last frame of the painting image in the painting image sequence by using a gray level co-occurrence matrix to obtain a texture feature map; Based on the Hough change algorithm, straight lines are extracted from the texture feature map to obtain line feature data; Segmenting the texture feature map based on a watershed algorithm to obtain a plurality of texture regions, and matching each of the texture regions with a preset pattern library to obtain pattern feature data; The line feature data is subjected to a smoothness analysis to obtain the line smoothness data, and the pattern feature data is subjected to a position analysis and a proportion analysis to obtain the balance data and the proportion data accordingly.

6. The method for generating a painting suggestion according to claim 4, characterized in that: The color matching features include the center value of the main color cluster and the proportion of the main color cluster; determining the color matching features of the last frame of the painting image in the painting image sequence includes: Obtaining the RGB value of each pixel in the last frame of the painting image in the painting image sequence by a pixel traversal method; Performing color space conversion on the RGB value to obtain the HSV value of each pixel; The HSV values ​​are clustered to obtain the center values ​​of the main color clusters and the proportions of the main color clusters.

7. The method for generating a painting suggestion according to claim 1, characterized in that: The drawing suggestions include suggestions for optimizing the direction of lines, suggestions for creative deformation of graphics, and suggestions for color matching.

8. The method for generating a painting suggestion according to claim 3, characterized in that: The painting feature also includes a brush type; the method also includes: Determining a theoretical brush type based on the painting scene; When the brush type is different from the theoretical brush type, a brush switching suggestion is generated.

9. An intelligent drawing assistance system, characterized in that: include; A camera, used to obtain a sequence of painting images; A host computer includes a memory and a processor, wherein: The memory is used to store programs; The processor is coupled to the memory and is used to execute the program stored in the memory to implement the steps of the method for generating a painting suggestion according to any one of claims 1 to 7; A projector is used for projecting and displaying the painting suggestions and the painting image sequences.

10. The intelligent drawing assistance system according to claim 9, characterized in that: The intelligent drawing assistance system also includes a multi-touch film; The multi-touch film is used to operate the painting image sequence in response to the current painter's gestures.

Images