Realism enhancement method for digital twin models based on rendering engine

Abstract

The present application relates to the field of image enhancement and rendering technology, and specifically to a method for enhancing the sense of reality of a digital twin model based on a rendering engine, the method comprising: obtaining a rendered image for virtual scene rendering, obtaining a rendered grayscale image, constructing a texture trend pair for each pixel, and selecting a trend distribution pair for each pixel; analyzing the difference between the trend distribution pair of each pixel and the texture trend pair of each pixel in a local neighborhood to obtain a structural distribution characteristic degree, clustering all pixels based on this, analyzing the difference in the texture trend pairs of all pixels in the cluster where each pixel is located to obtain the adjacent structural difference degree of each pixel, combining the discrete degree of the grayscale value of each pixel in the local neighborhood of each pixel to obtain the rendering blur of each pixel, using a guided graph filtering algorithm to obtain a filtered image, and rendering the filtered image. The present application can improve the rendering effect of a virtual scene and enhance the authenticity of the rendering.

Description

Technical Field

[0001] The present application relates to the field of image enhancement rendering technology, and specifically to a method for enhancing the realism of a digital twin model based on a rendering engine. Background Art

[0002] Digital twin technology, which maps the state of physical entities through virtual models, has gained widespread application in recent years, driven by the development of technologies such as big data, cloud computing, the Internet of Things, and artificial intelligence (AI), across sectors such as manufacturing, smart cities, healthcare, and agricultural technology. The core of digital twin technology lies in reflecting the state and operational status of physical entities in virtual models through real-time data collection and analysis, enabling simulation, prediction, and optimization.

[0003] However, the realization of digital twin models requires high-quality visual presentation, and rendering engines are key technologies for achieving this goal. As digital twin application scenarios expand, higher requirements are placed on rendering engines. Enhancing the realism of digital twin models through rendering engines has become a key direction in the current development of digital twin technology.

[0004] When rendering virtual models, traditional techniques typically pre-process images through guided image filtering (GUIDIF). However, because traditional GUIDIF typically uses a fixed cutoff value ε for image processing, excessively large values ​​can result in blurred images and poor rendering quality. Small cutoff values ​​can slow rendering. Since virtual reality technology operates in real time, images at every moment vary. A fixed cutoff value can conflict with the real-time nature of virtual model rendering and image quality requirements, leading to slow rendering and reduced realism. Summary of the Invention

[0005] In order to solve the above technical problems, the purpose of this application is to provide a method for enhancing the realism of a digital twin model based on a rendering engine. The technical solutions adopted are as follows:

[0006] The embodiment of the present application provides a method for enhancing the sense of reality of a digital twin model based on a rendering engine, comprising the following steps:

[0007] Obtaining a rendering image for virtual scene rendering, and obtaining a rendered grayscale image through grayscale processing;

[0008] By rendering the distribution characteristics of the grayscale values ​​of each pixel in the local neighborhood of each pixel in the grayscale image, the texture trend pair of each pixel is constructed, and the trend distribution pair of each pixel is selected according to the grayscale difference between each pixel and the local neighborhood pixels;

[0009] Analyze the difference between the trend distribution pair of each pixel point and the texture trend pair of each pixel point in the local neighborhood to obtain the structural distribution characteristic between each pixel point and each pixel point in the local neighborhood, and cluster all pixels based on the structural distribution characteristic. After clustering, obtain the neighborhood structural difference of each pixel point through the difference of the texture trend pairs of all pixels in the cluster where each pixel point is located;

[0010] The discrete degree of the grayscale value of each pixel in the local neighborhood of each pixel is analyzed, and the rendering blurriness of each pixel is obtained by combining the difference of the neighboring structure of each pixel. The rendering blurriness of all pixels is linearly mapped to obtain the mapping value corresponding to each pixel. The rendered image is filtered using the guided graph filtering algorithm, and the filtered image is rendered.

[0011] Preferably, the construction of the texture trend pair of each pixel point includes: for each pixel point, forming a texture trend pair of each pixel point with the grayscale values ​​of each pixel point and its eight neighboring pixels.

[0012] Preferably, determining the trend distribution pair of each pixel point includes: calculating the absolute value of the grayscale difference between each pixel point and its eight neighboring pixel points, and selecting the texture trend pair corresponding to the neighboring pixel point with the largest absolute value as the trend distribution pair of each pixel point.

[0013] Preferably, the method for calculating the structural distribution characteristic degree between each pixel point and each pixel point in the local neighborhood is:

[0014] Calculate the difference between the trend distribution pair of each pixel point and each texture trend pair of each pixel point in the eight neighborhoods to obtain the structural direction distribution degree between the trend distribution pair of each pixel point and each texture trend pair of each pixel point in the eight neighborhoods;

[0015] Then the structural distribution characteristic expression between each pixel and each pixel in the local neighborhood is: ; Represents the structural distribution characteristic between pixel u and the i-th pixel in its eight neighborhoods; It represents the set of structural direction distribution degrees between the trend distribution pair of pixel u and all texture trend pairs of the i-th pixel in its eight neighborhoods; min() represents the minimum value function.

[0016] Preferably, the calculation formula for the structural direction distribution degree between the trend distribution pair of each pixel point and each texture trend pair of each pixel point in the eight neighborhoods is: Where, It represents the structural direction distribution degree between the trend distribution pair of pixel u and the j-th texture trend pair of pixel i in its eight neighborhoods. 、 They represent the first and second elements of the trend distribution of pixel u respectively; 、 Represents the first and second elements of the j-th texture trend pair of the i-th pixel in the eight-neighborhood of the pixel u; || represents the absolute value.

[0017] Preferably, the process of clustering all pixel points further includes: forming a structural distribution feature sequence of each pixel point by the structural distribution feature degree between each pixel point and all pixel points in the eight-neighborhood area, taking the structural distribution feature sequence of all pixel points as the input of the clustering algorithm, and taking the difference in the structural distribution feature degree between any two pixel points as the metric distance of the clustering process to cluster the pixel points.

[0018] Preferably, the method for calculating the neighborhood structure difference of each pixel point is: Where, represents the neighborhood structure difference of pixel u; N represents the number of texture trend pairs in the texture trend set of pixel u; 、 They represent the first and second elements of the kth texture trend pair in the texture trend set of pixel u respectively;

[0019] The texture trend set is a set consisting of texture trend pairs of all pixels in the cluster where the pixel point is located.

[0020] Preferably, the corresponding calculation method for the rendering blurriness of each pixel is: Where, Indicates the rendering blur of pixel u; Represents the standard deviation of the grayscale values ​​of all pixels in the eight neighborhoods of pixel u; Indicates the difference of the neighboring structure of pixel u; Norm() is the normalization function; Represents a constant that prevents the denominator from being zero.

[0021] Preferably, the filtering of the rendered image by the guided map filtering algorithm further comprises: using the mapping value corresponding to each pixel point as a cutoff value in the filtering process of the rendered image by the guided filtering algorithm, so as to perform filtering processing on each pixel point in the rendered image.

[0022] Preferably, the grayscale processing adopts a weighted average method.

[0023] As can be seen from the above, the method for enhancing the realism of a digital twin model based on a rendering engine provided by this application has at least the following beneficial effects:

[0024] In order to solve the problems of slow rendering speed and low image quality in the existing technology due to the real-time and image quality requirements of virtual model rendering, this application analyzes the local features and structural distribution features of the pixels of the virtual scene image in real time to characterize the structural features of different objects in the virtual scene; by calculating the structural distribution feature degree and adjacent structural difference of the pixels, the boundaries and regions of the objects in the virtual scene are characterized, which is used to identify the boundaries between different texture structures in the image;

[0025] Furthermore, the rendering blurriness of the pixel points is obtained to determine the degree of blur at different locations in the rendered image, thereby maintaining the rendering image rate while ensuring the details of the rendering, resulting in a better rendering effect. Therefore, the present application avoids the problems of low rendering efficiency and weak rendering realism caused by the real-time and image quality requirements of the virtual model rendering method in the prior art, and can effectively improve the rendering quality and efficiency of the virtual model, making the rendering speed of the virtual model faster, the rendering effect better, and the rendering realism stronger. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present application or the prior art, the following is a brief introduction to the drawings required for use in the embodiments or the description of the prior art. 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 any creative work.

[0027] Figure 1 A flowchart of the steps of a method for enhancing the sense of reality of a digital twin model based on a rendering engine provided in this application;

[0028] Figure 2 Schematic diagram of constructing texture trend pairs from eight neighborhoods of pixels provided in this application. DETAILED DESCRIPTION

[0029] In order to further illustrate the technical means and effects adopted by this application to achieve the predetermined invention purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, describes in detail the specific implementation method, structure, features and effects of a digital twin model reality enhancement method based on a rendering engine proposed in this application. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0030] Unless otherwise specified and limited, terms such as "comprises", "includes" or any other variants thereof are intended to cover non-exclusive inclusion, so that a circuit structure, article or device comprising a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the presence of other identical elements in the article or device comprising the element. In addition, the term "and\or" used herein includes any and all combinations of one or more related listed items. All technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs.

[0031] The following describes in detail a specific scheme of a method for enhancing the realism of a digital twin model based on a rendering engine provided by this application with reference to the accompanying drawings.

[0032] See also Figure 1 , which shows a flowchart of a method for enhancing the sense of reality of a digital twin model based on a rendering engine provided by one embodiment of the present application, including the following steps:

[0033] Step 1: Obtain a rendering image for virtual scene rendering, and obtain a rendered grayscale image through grayscale processing.

[0034] First, the digital twin model's real-world scene is tracked in real time using infrared or ultrasonic technology, and video is captured through a camera to obtain a video frame image of the digital twin model, which is recorded as a rendered image. To improve the efficiency of image data processing, the rendered image is preprocessed. There are many existing image preprocessing methods, and in actual application scenarios, the implementer can set and select them. In this embodiment, the rendered image is converted into a grayscale image using the weighted averaging method, which is recorded as the rendered grayscale image. The calculation of the weighted averaging method is a well-known technology, and the specific calculation process is not repeated here.

[0035] Step 2: By rendering the distribution characteristics of the grayscale values ​​of each pixel in the local neighborhood of each pixel in the grayscale image, a texture trend pair of each pixel is constructed, and the trend distribution pair of each pixel is selected based on the grayscale difference between each pixel and the local neighborhood pixels.

[0036] To maintain high rendering quality in virtual scenes, it's crucial to maintain high detail in the rendered image. In reality, various substances aren't represented as single pixels in rendered images. The structures of the same substance typically occupy a specific area. The distribution of pixels within the same material region exhibits similar characteristics, and the intersections of different structures also exhibit similar characteristics.

[0037] For example, at the boundary intersection of different structures, the similarity of the pixels at the intersection does not change with the location of the intersection; similar features must be connected features. Therefore, in this embodiment, with each pixel of the rendered grayscale image as the center, the eight neighboring pixels of the center pixel are obtained. The grayscale values ​​of the center pixel and each of its eight neighboring pixels are combined to form a texture trend pair for the center pixel, and the eight texture trend pairs are combined to form a texture trend pair set.

[0038] Specifically, in this embodiment, the eight neighborhoods of pixel points are distributed to construct a schematic diagram of texture trend pairs as shown in FIG. Figure 2 As shown, Figure 2 The grayscale value of the central pixel is 0, and the grayscale values ​​of the eight neighboring pixels are 1, 2, 3, 4, 5, 6, 7, and 8 respectively. Figure 2 In the figure, the texture trend pairs of the central pixels are (0, 1), (0, 2), (0, 3), (0, 4), (0, 5), (0, 6), (0, 7), and (0, 8).

[0039] At this point, a set of texture trend pairs for each pixel in the rendered image is obtained, which is used to characterize the structural changes between the pixel and its surrounding pixels, so as to facilitate the identification of similar directions of the central pixel structure.

[0040] Furthermore, different structures have different grayscale values ​​in the rendered image. Therefore, the direction with the largest grayscale value difference is the direction of the different structural transformation. Therefore, for each central pixel, the absolute value of the grayscale value difference between the central pixel and its eight neighboring pixels is calculated, and the neighboring pixel with the largest absolute value is selected to form a trend distribution pair with the central pixel to characterize the characteristics of the structural change of the central pixel.

[0041] Step 3: Analyze the difference between the trend distribution pair of each pixel point and the texture trend pair of each pixel point in the local neighborhood to obtain the structural distribution characteristic between each pixel point and each pixel point in the local neighborhood, and cluster all pixels based on the structural distribution characteristic. After clustering, the difference in the texture trend pairs of all pixels in the cluster where each pixel point is located is used to obtain the neighborhood structural difference of each pixel point.

[0042] Furthermore, in this embodiment, taking the eight neighborhoods of any pixel point as an example, the difference between the trend distribution pairs of each pixel point and each texture trend pair of the pixel points in the eight neighborhoods is analyzed to obtain the structural direction distribution degree between the trend distribution pair of each pixel point and each texture trend pair of each pixel point in the eight neighborhoods. In this embodiment, the specific calculation relationship is: Where, It represents the structural direction distribution degree between the trend distribution pair of pixel u and the j-th texture trend pair of pixel i in its eight neighborhoods. 、 They represent the first and second elements of the trend distribution of pixel u respectively; 、 Represents the first and second elements of the j-th texture trend pair of the i-th pixel in the eight-neighborhood of the pixel u; || represents the absolute value.

[0043] Furthermore, based on the structural direction distribution degree between the trend distribution pair of each pixel point and all the texture trend pairs of each pixel point in its eight neighborhoods, the structural distribution characteristic degree between each pixel point and each pixel point in its eight neighborhoods is calculated. In this embodiment, the specific calculation relationship is: ;in, Represents the structural distribution characteristic between pixel u and the i-th pixel in its eight neighborhoods; It represents the set of structural direction distribution degrees between the trend distribution pair of pixel u and all texture trend pairs of the i-th pixel in its eight neighborhoods; min() represents the minimum value function.

[0044] When there is no noise in the rendered grayscale image, the pixels in the eight neighborhoods of the central pixel are of the same structural material, and there are other continuous pixels in the eight neighborhoods that have the same structure as the central pixel. Therefore, in the absence of noise, the smaller the difference between the texture distribution pair of the central pixel and the texture distribution pair of the remaining pixels, that is, The smaller the value, the minimum Therefore, the structural distribution characteristics between the eight neighborhood pixels and the central pixel are smaller, which is used to analyze the direction in which the neighborhood pixels and the central pixel are most similar in structure to evaluate whether the neighborhood pixels and the central pixel have the same texture structure.

[0045] In addition, when the trend distribution pair of the central pixel differs significantly from the texture trend pair of the eight neighboring pixels, it indicates that the central pixel is at the junction of two different structural materials. Since the junction of different materials is not necessarily a straight line but may also be curved, the texture trends of the pixels in multiple directions are used to calculate the structural distribution characteristic between the central pixel. When the local structural distribution of the pixels in the eight neighborhoods is similar to that of the central pixel, that is, the difference between the texture trend pair in the direction of the neighboring pixels and the trend distribution pair of the central pixel is smaller, the structural distribution characteristic between the central pixel and the neighboring pixels is smaller, thereby better distinguishing the structural distribution characteristics between pixels in the rendered grayscale image.

[0046] At this point, according to the above process of this embodiment, the structural distribution feature between each pixel point and each pixel point in its eight neighborhoods can be obtained. Further, the structural distribution feature between each pixel point and all pixels in the eight neighborhoods is formed into a structural distribution feature sequence of each pixel point.

[0047] Through the above steps, the structural distribution feature sequence of each pixel in the rendered grayscale image is obtained. From this, the neighborhood structural difference of each pixel in the rendered grayscale image is calculated. Taking any pixel as an example: when the central pixel is not at the boundary of the structure, the texture feature difference between the central pixel and the pixels of the same structure in its eight neighborhoods is small; when the central pixel is at the boundary of the structure, the texture feature difference between the central pixel and the pixels in its eight neighborhoods is large. Therefore, the structural distribution feature sequence of all pixels is used as the input of the K-means clustering algorithm, where the value of the necessary parameter cluster number K is 3, the structural distribution feature difference between pixels is used as the clustering metric distance, and the pixels are classified and output as three clusters. The calculation of the K-means clustering algorithm is a well-known technology, and the specific calculation process will not be repeated here.

[0048] Furthermore, for each pixel point in the cluster, the more pixels there are in the cluster, the more pixels are similar to the local structural distribution characteristics of the pixel point. If the pixel point is a structural edge point, the higher the complexity of representing the pixel point, the smaller the filtering degree of the pixel point position is to ensure the rendering effect. For each pixel point, a texture tendency set consisting of texture tendency pairs corresponding to all pixels in the cluster where the pixel point is located is obtained, which is used to represent the set of pixels with similar texture characteristics to the pixel point, recorded as According to the differences in texture tendency pairs of all pixels in the cluster where each pixel is located, the neighborhood structure difference of each pixel is calculated. In this embodiment, the specific calculation relationship is: Where, represents the neighborhood structure difference of pixel u; N represents the number of texture trend pairs in the texture trend set of pixel u; 、 They respectively represent the first and second elements of the kth texture trend pair in the texture trend set of pixel u.

[0049] When a pixel is at the edge of a structure, the greater the difference between the pixel and other structural pixels, that is, the absolute value of the grayscale value difference between the pixels. The larger the value is, the greater the degree of detail retention for the pixel should be during the filtering process of the scene image. Therefore, the greater the difference in the neighboring structure calculated based on the difference in the texture features of the eight neighborhoods of the pixel is, so as to ensure that more detail information is retained when the virtual scene image is filtered and rendered; conversely, the central pixel is inside the structure, that is, there is less detail information in the local neighborhood of the pixel, and the texture structure is more uniform, which can retain lower detail attributes, thereby reducing the time and energy consumption in the rendering process.

[0050] Thus, according to the above process of this embodiment, the neighborhood structural difference of each pixel in the rendered grayscale image can be obtained, which is used to characterize the structural change in the neighborhood of each pixel.

[0051] Step 4: Analyze the discrete degree of the grayscale value of each pixel in the local neighborhood of each pixel, and combine the neighborhood structure difference of each pixel to obtain the rendering blur of each pixel, and perform linear mapping on the rendering blur of all pixels to obtain the mapping value corresponding to each pixel. Use the guided graph filtering algorithm to filter the rendered image, and render the filtered image.

[0052] Regarding the denoising and enhancement of virtual scene images, during the rendering process of the virtual scene, the higher the detail attributes between different structures in the virtual scene image, the better the rendering effect of the virtual scene will be after rendering. However, images with high detail attributes require a longer rendering time, which makes the real-time effect of virtual reality technology poor and brings higher energy consumption. Therefore, it is necessary to denoise and enhance the image, retain the detail attributes at the junction of different structures in the image, and retain the detail attributes within the same structure, so that the virtual reality technology can render the virtual scene image faster during the rendering process.

[0053] It can be understood that the neighborhood structure difference of each pixel point represents the degree of structural change at each position of the rendered grayscale image. The greater the neighborhood structure difference of each pixel point, the greater the possibility that the pixel point at the corresponding position in the rendered grayscale image is at the edge of the structure. Greater detail attributes should be retained for this part. After virtual reality technology is used to render this position, there is a clear layering between different structures.

[0054] Based on the above analysis, the rendering blurriness of each pixel in the rendered grayscale image is calculated by the grayscale dispersion within the local neighborhood of each pixel in the rendered grayscale image and the difference in the neighboring structure of each pixel. In this embodiment, the specific calculation relationship is: Where, Indicates the rendering blur of pixel u; Represents the standard deviation of the grayscale values ​​of all pixels in the eight neighborhoods of pixel u; Indicates the difference of the neighboring structure of pixel u; Norm() is the normalization function; Indicates a constant to prevent the denominator from being zero, with a value ranging from 0 to 0.1. In this embodiment, to prevent the denominator from being zero, the value is 0.001.

[0055] When the local distribution of pixels in the rendered grayscale image has obvious differences, it indicates that the location corresponds to a different texture structure, so the standard deviation of the grayscale values ​​of all pixels in the eight neighborhoods of the pixel point is And the difference of the neighboring structure of the pixel The larger the value is, the more detailed attributes should be retained at the pixel location in the rendered grayscale image to ensure better rendering effect through virtual reality technology. Therefore, the rendering fuzziness of the pixel is The smaller it is, the better the rendering effect of virtual reality technology on virtual scenes.

[0056] Furthermore, in this embodiment, the rendering blur of each pixel in the rendered image is used to form a rendered blurred image. In image processing, when the local detail attribute is greater, it means that the local changes and features in the image are richer and more complex. In this case, in order to better preserve these details and features, the truncation value ε should be smaller. Therefore, based on the above analysis, all values ​​in the rendered blurred image are mapped to , the mapping value corresponding to each pixel is obtained, and the mapping values ​​of all pixels constitute a mapping image. Among them, the calculation of linear mapping is a well-known technology, and the specific calculation process is not repeated here.

[0057] The mapped image obtained in the above process replaces the cutoff value in the guided image filtering, improving the guided image filtering. Specifically, the mapped value of each pixel is used as the cutoff value in the guided image filtering process to filter each pixel. Furthermore, the rendered image is used as the input for the improved guided image filtering. In this embodiment, the required parameter filter window r is set to 13, and the output is the filtered and enhanced image.

[0058] Furthermore, the filtered and enhanced image is transmitted to a real-time rendering engine. Combining rasterization and ray tracing technology, the GPU parallel computing capability is used to achieve high-speed rendering of multiple frames per second. Real-time rendering engines such as Unreal Engine and Unity. The rendering process is a common technology in the field of video rendering, and the specific operation steps will not be repeated here.

[0059] It should be understood that the order in which the embodiments of the present application are presented is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. Furthermore, the foregoing descriptions of specific embodiments of this specification are provided. Furthermore, the processes depicted in the accompanying drawings do not necessarily require the specific order or sequential sequence shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0060] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments.

[0061] The above content is only an implementation method of the present application and is not intended to limit the scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the scope of protection of the present application.

Claims

1. A method for enhancing the sense of reality of a digital twin model based on a rendering engine, characterized in that: The following steps are involved: Obtaining a rendering image for virtual scene rendering, and obtaining a rendered grayscale image through grayscale processing; By rendering the distribution characteristics of the grayscale values ​​of each pixel in the local neighborhood of each pixel in the grayscale image, the grayscale values ​​of each pixel and its eight neighboring pixels are respectively formed into texture trend pairs of each pixel, and the absolute value of the grayscale difference between each pixel and its eight neighboring pixels is calculated. The texture trend pair corresponding to the neighboring pixel with the largest absolute value is selected as the trend distribution pair of each pixel; Calculate the difference between the trend distribution pair of each pixel point and each texture trend pair of the pixel points in the eight neighborhoods to obtain the structural direction distribution degree between the trend distribution pair of each pixel point and each texture trend pair of each pixel point in the eight neighborhoods; take the minimum value of the set of structural direction distribution degrees between the trend distribution pair of the pixel point and all texture trend pairs of any pixel point in its eight neighborhoods as the structural distribution characteristic degree between each pixel point and each pixel point in the local neighborhood, and cluster all pixels based on the structural distribution characteristic degree. After clustering, obtain the neighborhood structural difference degree of each pixel point through the difference of the texture trend pairs of all pixels in the cluster where each pixel point is located. Analyze the discrete degree of the grayscale value of each pixel in the local neighborhood of each pixel, and combine it with the difference of the neighboring structure of each pixel to obtain the rendering blur of each pixel, and perform linear mapping on the rendering blur of all pixels to obtain the mapping value corresponding to each pixel. The mapping value corresponding to each pixel is used as the cutoff value in the filtering process of the rendered image by the guided filtering algorithm, the rendered image is filtered, and the filtered image is rendered; The calculation method of the neighborhood structure difference of each pixel point is: Where, represents the neighborhood structure difference of pixel u; N represents the number of texture trend pairs in the texture trend set of pixel u; 、 They represent the first and second elements of the kth texture trend pair in the texture trend set of pixel u respectively; the texture trend set is the set composed of the texture trend pairs of all pixels in the cluster where the pixel is located; The corresponding calculation method for the rendering blurriness of each pixel is: Where, Indicates the rendering blur of pixel u; Represents the standard deviation of the grayscale values ​​of all pixels in the eight neighborhoods of pixel u; Norm() is the normalization function; Represents a constant that prevents the denominator from being zero.

2. The method for enhancing the sense of reality of a digital twin model based on a rendering engine according to claim 1, characterized in that: The calculation formula for the structural direction distribution degree between the trend distribution pair of each pixel point and each texture trend pair of each pixel point in the eight neighborhoods is: Where, It represents the structural direction distribution degree between the trend distribution pair of pixel u and the j-th texture trend pair of pixel i in its eight neighborhoods. 、 They represent the first and second elements of the trend distribution of pixel u respectively; 、 Represents the first and second elements of the j-th texture trend pair of the i-th pixel in the eight-neighborhood of the pixel u; || represents the absolute value.

3. The method for enhancing the sense of reality of a digital twin model based on a rendering engine according to claim 1, characterized in that: The process of clustering all pixel points further includes: forming a structural distribution feature sequence of each pixel point by the structural distribution feature degree between each pixel point and all pixel points in the eight-neighborhood area, taking the structural distribution feature sequence of all pixel points as the input of the clustering algorithm, and taking the difference in the structural distribution feature degree between any two pixel points as the metric distance of the clustering process to cluster the pixel points.

4. The method for enhancing the sense of reality of a digital twin model based on a rendering engine according to claim 1, wherein: The grayscale processing adopts a weighted average method.

Images