Analyzing and processing method based on skin texture graphic data

Abstract

The invention discloses an analysis processing method based on epidermis texture graphic data, and relates to the technical field of epidermis texture data analysis, and the method comprises the steps: carrying out the three-dimensional scanning of a target epidermis to obtain an initial texture model, connecting three adjacent surface data points based on a Delaunay triangulation algorithm to form a triangulation analysis unit, and carrying out the triangulation analysis of the target epidermis. Calculating a normal vector included angle standard deviation of adjacent triangular patches in the triangular analysis unit where the texture reference point is located as a texture direction change rate, judging whether the texture direction change rate is greater than or equal to a preset threshold value, and performing skin curvature feature extraction on the current triangular analysis unit with first analysis intensity, according to the first analysis intensity or the second analysis intensity, epidermal frequency feature extraction is carried out on the current triangular analysis unit, so that different analysis intensities are intelligently switched, the key direction of feature extraction is flexibly adjusted, it is ensured that high-precision analysis is obtained in a key complex area, meanwhile, the overall calculation burden is effectively reduced, and the processing efficiency is improved.

Description

Technical Field

[0001] The present application relates to the technical field of skin texture data analysis, and in particular to an analysis and processing method based on skin texture graphic data. Background Art

[0002] Existing epidermal texture analysis technologies face a core challenge: striking a balance between processing efficiency and analysis accuracy. Traditional methods typically employ a uniform analysis intensity and strategy across the entire epidermal region. This uniform approach results in unnecessary high computational resources and data sampling density when processing areas with relatively flat texture features and minimal directional variation, resulting in wasted resources and inefficient processing. However, this fixed analysis intensity is often insufficient in areas with drastic and directional variations in texture features.

[0003] Due to insufficient sampling point density or insufficient feature extraction depth, key detail information may be lost, making it difficult to accurately capture and quantify indicators that are critical to reflecting subtle epidermal pathology or physical properties, such as significant curvature extremes or high-frequency texture energy.

[0004] A key drawback of existing technologies is their lack of intelligent recognition of the complexity of localized epidermal textures, particularly their directional consistency. This inability to dynamically adjust analysis strategies and resource allocation based on the actual complexity of the textures is crucial. Therefore, a technical solution that can automatically identify the degree of directional variation in localized epidermal textures is urgently needed. Summary of the Invention

[0005] In an exemplary embodiment of the present application, an analysis and processing method based on epidermal texture graphic data is provided to intelligently switch between different analysis intensities and flexibly adjust the focus of feature extraction, thereby ensuring high-precision analysis in key complex areas while effectively reducing the overall computational burden and improving processing efficiency.

[0006] This application provides an analysis and processing method based on skin texture graphic data, comprising: Performing a three-dimensional scan of the target epidermis to obtain an initial texture model, wherein the initial texture model is composed of a plurality of surface data points, connecting three adjacent surface data points based on a Delaunay triangulation algorithm to form a triangular analysis unit, and determining a texture reference point at the geometric center of gravity of each triangular analysis unit; Calculating the standard deviation of the normal vector angles of adjacent triangles in the triangular analysis unit where the texture reference point is located as the texture direction change rate, and determining whether the texture direction change rate is greater than or equal to a preset threshold; If the texture direction change rate is greater than or equal to the preset threshold, performing skin curvature feature extraction on the current triangulation analysis unit with a first analysis intensity; If the texture direction change rate is less than the preset threshold, skin frequency feature extraction is performed on the current triangular analysis unit with a second analysis intensity to extract the texture spatial frequency spectrum energy value of the triangular analysis unit.

[0007] Furthermore, the amount of feature extraction data corresponding to the first analysis intensity is 1.8-2.2 times that of the second analysis intensity.

[0008] Furthermore, the skin curvature feature extraction specifically includes: extracting the Gaussian curvature mean and maximum curvature value within the triangular analysis unit.

[0009] Furthermore, the epidermal curvature feature extraction includes: calculating the Gaussian curvature of all sampling points in the triangular analysis unit, taking the average of the absolute values of the Gaussian curvature as the Gaussian curvature mean, and taking the maximum absolute value of the curvature in the triangular analysis unit as the maximum curvature value.

[0010] Furthermore, the skin frequency feature extraction includes: performing Fourier transform on the triangular analysis unit, and extracting the energy integral value within the frequency band of 0.1 Hz-10 Hz as the texture spatial frequency spectrum energy value.

[0011] Furthermore, determining the texture direction change rate includes: obtaining all adjacent triangular facets of the current triangular analysis unit, calculating the angle between the normal vectors of each adjacent triangular facet and the normal vector of the current triangular analysis unit, and determining the texture direction change rate based on the discrete degree of the angle.

[0012] Furthermore, the sampling point density corresponding to the first analysis intensity is greater than the sampling point density corresponding to the second analysis intensity.

[0013] The embodiments of the present application have the following beneficial effects: the technical solution constructs an adaptive epidermal texture analysis mechanism, which effectively optimizes the balance between processing efficiency and feature resolution accuracy. The core is to use the texture direction change rate as a dynamic discrimination criterion and intelligently adjust the analysis strategy based on the triangular mesh unit. By calculating the standard deviation of the angle between the normal vectors of adjacent triangles in the unit, the system can accurately distinguish between areas with drastic changes in texture direction and relatively gentle areas. For complex texture areas with significant directional changes, the system adopts a high-intensity analysis mode. By extracting key geometric features such as the mean and maximum curvature values of Gaussian curvature, it deepens the ability to depict complex forms such as epidermal wrinkles and scars, ensuring accurate capture and quantification of key details. For homogeneous texture areas with gentle directional changes, the system enables a low-intensity analysis mode, extracts the texture spatial frequency spectrum energy value of a specific frequency band through Fourier transform, efficiently characterizes the periodic distribution characteristics of the texture, significantly reduces unnecessary computational overhead, and automatically switches the dual-mode processing of analysis intensity and focus according to the local texture complexity, overcoming the resource mismatch problem caused by unified processing of traditional methods. It achieves deeper feature analysis in complex areas and avoids redundant calculations in simple areas, significantly reducing the overall consumption of computing resources, improving the intelligence level and efficiency of surface texture feature extraction, and providing a more efficient and detailed analysis basis for applications such as material surface quality inspection. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the technical solutions in the embodiments of the present application, 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 application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0015] Figure 1 The flowchart of an analysis and processing method based on skin texture graphic data provided by an embodiment of the present application is exemplified. DETAILED DESCRIPTION

[0016] In order to make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application.

[0017] To further illustrate the technical solutions provided by the embodiments of the present application, the following is a detailed description of the technical solutions in conjunction with the accompanying drawings and specific implementation methods. Although the embodiments of the present application provide the method operation steps as shown in the following embodiments or drawings, more or fewer operation steps may be included in the method based on routine or no creative work. In steps where there is no necessary causal relationship logically, the execution order of these steps is not limited to the execution order provided in the embodiments of the present application.

[0018] refer to Figure 1 As shown, the present application provides an analysis and processing method based on skin texture graphic data, comprising: S1: Perform a 3D scan of the target epidermis to obtain an initial texture model. The initial texture model is composed of multiple surface data points. Three adjacent surface data points are connected based on the Delaunay triangulation algorithm to form a triangular analysis unit. The texture reference point is determined at the geometric center of gravity of each triangular analysis unit.

[0019] This method uses 3D scanning to obtain an initial texture model as the basis for data processing. Its purpose is to transform the spatial topography of the target epidermis into a digital model that can be quantified and analyzed. By acquiring densely distributed surface data points, the spatial coordinate information of the epidermal microstructure is fully recorded, providing raw data support for subsequent feature extraction.

[0020] The Delaunay triangulation algorithm connects adjacent surface data points to form triangular analysis units, primarily to construct a topologically stable analysis mesh. This algorithm avoids narrow triangles by maximizing the minimum internal angle principle, ensuring the regularity of the mesh geometry.

[0021] The establishment of a triangular mesh structure effectively characterizes the local continuity of surface textures, providing a foundational topological framework for calculating geometric features such as normals and curvature. In particular, this structure provides a clear mathematical definition of the spatial relationship between adjacent facets, creating the necessary conditions for the subsequent calculation of the rate of change of texture directions.

[0022] A texture reference point is set at the geometric centroid of each triangular analysis unit. Its core purpose is to establish a benchmark position for local feature analysis. The geometric centroid is stable and unaffected by vertex distribution deviations, providing a more balanced representation of the overall geometric properties of the triangular unit. This reference point serves as a bridge between microscopic data points and macroscopic texture features, allowing the system to radiate and calculate the geometric relationships between the unit and its neighbors, centered on this point.

[0023] This setting not only ensures the regional representativeness of the normal vector angle calculation, but also provides a unified spatial coordinate benchmark for the subsequent differentiated extraction of curvature features and frequency features, thereby ensuring the comparability and spatial consistency of feature data under different analysis modes.

[0024] S2: Calculate the standard deviation of the normal vector angles of adjacent triangles in the triangular analysis unit where the texture reference point is located as the texture direction change rate, and determine whether the texture direction change rate is greater than or equal to a preset threshold.

[0025] Determining the texture direction change rate includes: obtaining all adjacent triangles of the current triangular analysis unit, calculating the angle between the normal vectors of each adjacent triangle and the normal vector of the current triangular analysis unit, and determining the texture direction change rate based on the discrete degree of the angle.

[0026] This parameter accurately characterizes the degree of orientation consistency of the surface microstructure of a triangular analysis unit by calculating the standard deviation of the angles between the normal vectors of adjacent triangles within that unit. This mathematical definition effectively captures the discrete nature of texture orientation at the microscale, providing an objective basis for the selection of subsequent differentiation analysis strategies.

[0027] The texture direction change rate plays a pivotal role in switching analysis modes within the overall solution. When this value exceeds a preset threshold, it indicates a significant change in texture direction in the current region, corresponding to highly complex features such as epidermal wrinkles, scars, or the edges of material defects.

[0028] At this point, the system automatically triggers high-intensity curvature analysis mode, increasing sampling density and data extraction to deeply explore key geometric parameters such as the mean and maximum Gaussian curvature values, ensuring accurate analysis of complex topography. Conversely, when this value falls below the threshold, it is determined to be a homogeneous area with high texture directional consistency, and the system switches to low-intensity frequency analysis mode, efficiently extracting the texture's spatial frequency spectrum energy value through Fourier transform, avoiding resource waste caused by excessive calculations.

[0029] This dynamic discrimination mechanism, based on objective mathematical metrics, successfully overcomes the resource mismatch problem caused by the traditional method of using uniform analysis intensity. This parameter enables the system to intelligently distinguish the essential characteristic differences between different regions of the epidermis, enhancing feature extraction depth in key complex areas and reducing the computational load in simple, homogeneous areas. This achieves an overall coordinated improvement in the optimization of computing resources and feature analysis accuracy, providing more efficient technical support for skin pathology diagnosis and industrial surface inspection.

[0030] S3: If the texture direction change rate is greater than or equal to the preset threshold, performing skin curvature feature extraction on the current triangular analysis unit with a first analysis intensity.

[0031] The epidermal curvature feature extraction specifically includes: extracting the Gaussian curvature mean and maximum curvature value within the triangular analysis unit.

[0032] The epidermal curvature feature extraction includes: calculating the Gaussian curvature of all sampling points in the triangular analysis unit, taking the average of the absolute values of the Gaussian curvature as the Gaussian curvature mean, and taking the maximum absolute value of the curvature in the triangular analysis unit as the maximum curvature value.

[0033] This method uses the extraction of mean and maximum Gaussian curvature as the core of epidermal curvature analysis, aiming to construct a multi-dimensional micro-deformation quantification system. Gaussian curvature, a key indicator of the intrinsic geometric properties of a surface, can effectively characterize the overall curvature characteristics of the epidermal texture by calculating the absolute average of the Gaussian curvature at all sampling points within a triangulated analysis unit.

[0034] This parameter is highly sensitive to the distribution density of concave and convex structures and can quantitatively reflect regional deformation characteristics such as pore clustering and micro-wrinkle distribution. The maximum curvature value, by capturing the extreme points of the absolute curvature within a triangular unit, accurately locates microstructures with the most significant local deformation, such as scar vertices and material crack tips, which are diagnostically important mutation features. The two complement each other in their mathematical definitions: the mean Gaussian curvature reveals the overall deformation trend of the region, while the maximum curvature value identifies local anomalies and extreme values, together forming a complete description of complex epidermal geometric deformation.

[0035] This parameter combination has clear diagnostic significance within the technical solution. In areas where texture direction changes dramatically, the mean Gaussian curvature can effectively distinguish the distribution ratios of raised and recessed structures, providing data support for skin elasticity assessment. The maximum curvature value can identify tiny pathological protrusions or material defect vertices that are difficult to discern with the naked eye, thereby increasing the sensitivity of early lesion identification. Especially for scar tissue assessment, this parameter combination can simultaneously quantify the overall shrinkage degree and edge irregularity of the scar area, providing a precise quantitative basis for treatment.

[0036] Compared to frequency feature analysis, curvature parameter extraction directly correlates to the true nature of the epidermal three-dimensional topological deformation. In complex areas with high directional discreteness, curvature feature extraction overcomes the Fourier transform's inadequate representation of nonlinear deformations, ensuring the complete preservation of pathologically relevant geometric features. This differential geometry-based feature selection mechanism enables the technical solution to develop an analysis path that strictly matches the physical nature of textures in key areas, establishing a reliable data foundation for skin health diagnosis and industrial surface defect detection.

[0037] S4: If the texture direction change rate is less than the preset threshold, performing skin frequency feature extraction on the current triangular analysis unit with a second analysis intensity to extract the texture spatial frequency spectrum energy value of the triangular analysis unit.

[0038] The skin frequency feature extraction includes: performing Fourier transform on the triangular analysis unit, and extracting the energy integral value within the frequency band of 0.1 Hz-10 Hz as the texture spatial frequency spectrum energy value.

[0039] The amount of feature extraction data corresponding to the first analysis intensity is 1.8-2.2 times that of the second analysis intensity, and the sampling point density corresponding to the first analysis intensity is greater than the sampling point density corresponding to the second analysis intensity.

[0040] When the rate of change in texture direction exceeds a preset threshold, the system automatically triggers the first analysis intensity mode to perform curvature feature extraction. This high-intensity analysis increases sampling point density and data extraction, focusing on calculating the mean and maximum Gaussian curvature values within the triangular analysis unit, deeply analyzing the microscopic geometric features of complex deformation areas such as epidermal wrinkles and scars.

[0041] This targeted processing ensures sufficient feature resolution depth in sensitive areas where texture direction changes drastically, effectively capturing critical curvature extremes that characterize pathological conditions or material defects.

[0042] When the texture direction change rate falls below a preset threshold, the system switches to the second analysis intensity mode to perform frequency feature extraction. This mode extracts the texture spatial frequency spectrum energy value within the 0.1-10 Hz frequency band by performing a Fourier transform on the triangular analysis unit.

[0043] In homogeneous areas with high texture directional consistency, this low-intensity analysis mode can efficiently characterize the periodic distribution of textures, avoid redundant curvature calculations in flat areas, and significantly reduce the data processing load.

[0044] This dual-mode synergy forms a precise technical closed loop within the overall solution: through the objective quantitative indicator of texture directional change rate, it intelligently identifies the intrinsic characteristic differences of different areas of the epidermis; in areas of directional mutation, it strengthens the in-depth exploration of geometric deformation features, and in areas of directional stability, it focuses on the efficient capture of periodic textures.

[0045] It not only effectively solves the resource mismatch problem under the traditional homogenization processing mode, but also realizes the unified optimization of computing efficiency and analytical accuracy at the system level through the precise matching of feature extraction strategy and essential properties of texture, providing a technical foundation that is both economical and reliable for dermatological diagnosis and industrial product surface inspection.

[0046] The first analysis intensity primarily targets complex areas with dramatic variations in texture direction. By increasing sampling point density and data extraction, it enhances the ability to deeply analyze epidermal micro-geometric deformations. In this mode, the system focuses on calculating the mean and maximum Gaussian curvature values within triangular analysis units, accurately capturing the steep curvature changes that characterize the edges of wrinkles, scars, or material defects. This high-intensity processing strategy ensures that geometric deformation data reflecting pathological conditions or structural abnormalities is fully captured in critical areas with high texture direction dispersion, providing a reliable basis for precise diagnosis.

[0047] The second analysis intensity targets homogeneous areas with high texture directional consistency, employing a streamlined and efficient feature extraction path. This mode directly characterizes the periodic distribution of epidermal texture by applying a Fourier transform to the triangular analysis unit, extracting the energy values of the texture's spatial frequency spectrum within the 0.1-10Hz frequency band. This low-intensity processing avoids redundant geometric calculations in flat areas, significantly reducing the computational load. Furthermore, frequency-domain energy analysis effectively captures repetitive microtexture features that are invisible to the naked eye.

[0048] The differentiated configuration of two analysis intensities forms a precise technical closed loop within the overall solution: the first ensures deep feature resolution in complex areas, while the second improves processing efficiency in homogeneous areas. These two levels are intelligently switched through a threshold for texture direction change rate, ensuring that computing resources are dynamically matched to regional complexity.

[0049] This mechanism not only overcomes the resource mismatch problem caused by traditional homogenized analysis models, but also achieves an organic unity of analytical accuracy and computational efficiency at the system level by optimizing the extraction strategy of curvature and frequency features. This synergistic effect provides a technical foundation for both economical and reliable dermatological diagnosis and industrial surface testing.

[0050] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the present application. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0051] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0052] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0053] Obviously, those skilled in the art may make various changes and modifications to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is intended to include these modifications and variations.

Claims

1. A method for analyzing and processing skin texture graphic data, characterized in that: include: Performing a three-dimensional scan of the target epidermis to obtain an initial texture model, wherein the initial texture model is composed of a plurality of surface data points, connecting three adjacent surface data points based on a Delaunay triangulation algorithm to form a triangular analysis unit, and determining a texture reference point at the geometric center of gravity of each triangular analysis unit; Calculating the standard deviation of the normal vector angles of adjacent triangles in the triangular analysis unit where the texture reference point is located as the texture direction change rate, and determining whether the texture direction change rate is greater than or equal to a preset threshold; If the texture direction change rate is greater than or equal to the preset threshold, performing skin curvature feature extraction on the current triangulation analysis unit with a first analysis intensity; If the texture direction change rate is less than the preset threshold, skin frequency feature extraction is performed on the current triangular analysis unit with a second analysis intensity to extract the texture spatial frequency spectrum energy value of the triangular analysis unit.

2. The method according to claim 1, characterized in that The amount of feature extraction data corresponding to the first analysis intensity is 1.8-2.2 times that of the second analysis intensity.

3. The method according to claim 1, characterized in that The skin curvature feature extraction specifically includes: extracting the Gaussian curvature mean and maximum curvature value within the triangular analysis unit.

4. The method according to claim 3, characterized in that The epidermal curvature feature extraction includes: calculating the Gaussian curvature of all sampling points in the triangular analysis unit, taking the average of the absolute values of the Gaussian curvature as the Gaussian curvature mean, and taking the maximum absolute value of the curvature in the triangular analysis unit as the maximum curvature value.

5. The method according to claim 1, wherein The skin frequency feature extraction includes: performing Fourier transform on the triangular analysis unit, and extracting the energy integral value within the frequency band of 0.1 Hz to 10 Hz as the texture spatial frequency spectrum energy value.

6. The method according to claim 1, characterized in that Determining the texture direction change rate includes: obtaining all adjacent triangles of the current triangular analysis unit, calculating the angle between the normal vectors of each adjacent triangle and the normal vector of the current triangular analysis unit, and determining the texture direction change rate based on the discrete degree of the angle.

7. The method according to claim 1, characterized in that The sampling point density corresponding to the first analysis intensity is greater than the sampling point density corresponding to the second analysis intensity.

Images