Three-dimensional digital model interactive display method for high-cold and high-altitude area

Abstract

The invention relates to the technical field of remote sensing image enhancement, and provides a three-dimensional digital model interactive display method for a high-cold and high-altitude area, and the method comprises the steps: collecting a remote sensing image, point cloud data and geographic information system data of the high-cold and high-altitude area; respectively acquiring a measurement distance between each pixel point and each seed point in the remote sensing image, and dividing the remote sensing image into different landform areas; calculating a filtering weight of each pixel point in the same landform area, determining a correction illumination component of the landform area in combination with pixel values of all pixel points in the landform area, and obtaining a remote sensing enhanced image according to the correction illumination component of the landform area of the remote sensing image; and according to the remote sensing enhanced image, the point cloud data and the geographic information system data, three-dimensional digital model interactive display of the high-cold and high-altitude area is realized. According to the invention, the quality of the three-dimensional digital model in the high-cold and high-altitude area can be improved.

Description

Technical Field

[0001] This application relates to the field of remote sensing image enhancement technology, specifically to an interactive display method for three-dimensional digital models of high-altitude and cold regions. Background Technology

[0002] High-altitude and cold regions have complex terrains. Three-dimensional digital models of these regions can visually represent their topography, aiding in infrastructure planning and construction, geological disaster prevention, ecological environment monitoring and protection, and providing data support for research activities such as mountain uplift and glacier changes. However, snow cover and mountain shadows in these regions often lead to abnormal reflective points and mountain shadows in remote sensing images, resulting in uneven brightness, low contrast, and loss of regional topographic information. Three-dimensional digital models of high-altitude and cold regions built from remote sensing images with this information loss cannot accurately match the topographic information and therefore cannot accurately represent the terrain. Therefore, image enhancement is needed for remote sensing images of high-altitude and cold regions to avoid color distortion caused by strong reflective points and shadowed areas, thereby improving the quality of three-dimensional digital models of these regions.

[0003] The Retinex image enhancement algorithm can be used to enhance remote sensing images of high-altitude and cold regions. However, the Retinex algorithm cannot distinguish between noise and details such as image edges in remote sensing images, resulting in noticeable edge blurring and loss of detail in the enhanced images. Furthermore, 3D digital models of high-altitude and cold regions built from remote sensing images enhanced by the Retinex algorithm cannot accurately represent the terrain of these areas. Summary of the Invention

[0004] This application provides an interactive display method for three-dimensional digital models of high-altitude and cold regions to solve the problem of inaccurate three-dimensional digital models caused by the loss of remote sensing image information in these regions. The specific technical solution adopted is as follows: One embodiment of this application provides a method for interactively displaying a three-dimensional digital model of a high-altitude, cold region, the method comprising the following steps: Collect remote sensing images, point cloud data, and geographic information system data from high-altitude and cold regions; The SLIC superpixel segmentation algorithm is used to divide the remote sensing image. The positions of all seed points, the color distance and spatial distance between each pixel and each seed point in the remote sensing image are obtained. Based on the difference in gradient values, color distance and spatial distance between the pixels and seed points in the remote sensing image, as well as the pixel value of each seed point, the metric distance between each pixel and each seed point in the remote sensing image is obtained. Based on the metric distance between the pixels and seed points in the remote sensing image, the remote sensing image is divided into different landform regions. Based on the differences in pixel values, gradient values, and positions between pixels within the same topographic region in the remote sensing image and their eight neighboring pixels, the filtering weight of each pixel within the same topographic region is calculated. Based on the pixel values ​​and filtering weights of all pixels within the topographic region, the corrected illumination component of the topographic region is determined. Based on the corrected illumination component of the topographic region in the remote sensing image, the remote sensing enhanced image is obtained. Based on remote sensing enhanced images, point cloud data, and geographic information system data, a three-dimensional digital model of a high-altitude and cold region is constructed to enable interactive display of the three-dimensional digital model of the region.

[0005] Furthermore, the specific method for obtaining the metric distance between each pixel and each seed point in the remote sensing image based on the difference in gradient values, color distance, and spatial distance between pixels and seed points in the remote sensing image, as well as the pixel value of each seed point, includes the following: The negative correlation processing result of the mean pixel values ​​of seed points in the R, G, and B channels of the remote sensing image is recorded as the first weight of the seed points in the remote sensing image. Based on the difference in gradient values, color distance, and spatial distance between pixels and seed points in the remote sensing image, as well as the first weight of the seed points, the metric distance between each pixel and each seed point in the remote sensing image is obtained.

[0006] Furthermore, the formula for calculating the metric distance between each pixel and each seed point in the remote sensing image is: in, Represents pixels in a remote sensing image With seed point The distance between them; Represents seed points in remote sensing images The first weight; Represents pixels in a remote sensing image With seed point Color distance between them; Represents pixels in a remote sensing image With seed point Spatial distance between them; This indicates the preset second weight; Represents seed points in remote sensing images The number of pixels contained within the superpixel block; Represents pixels in a remote sensing image The gradient value; Represents seed points in remote sensing images The gradient value.

[0007] Furthermore, the specific method for dividing the remote sensing image into different landform regions based on the metric distance between pixels and seed points in the remote sensing image includes: The distance between a pixel and a seed point in the remote sensing image is used as the distance between the pixel and the seed point. The remote sensing image is divided using the SLIC superpixel segmentation algorithm to obtain superpixel regions, and all superpixel regions are recorded as terrain regions.

[0008] Furthermore, the specific method for calculating the filtering weight of each pixel within the same topographic region based on the pixel values, gradient values, and positional differences between pixels within the same topographic region and their eight neighboring pixels in the remote sensing image includes: Calculate the topographic region in the remote sensing image The internal coordinates are The third smoothing coefficient for each pixel is calculated as follows: in, Representing the topographic region in a remote sensing image The internal coordinates are The third smoothing coefficient of the pixels; This indicates the preset first parameter; This indicates the preset second parameter; Representing the topographic region in a remote sensing image The internal coordinates are The gradient value of the pixel; Based on the pixel values, gradient values, and positional differences between pixels within the same geomorphic region and pixels within their eight-neighborhood in the remote sensing image, as well as the third smoothing coefficient of the pixels, the third weight of pixels within the same geomorphic region is calculated. Based on the third weights of all pixels within the same topographic region and all pixels in their eight neighborhoods, the filtering weights of each pixel within the same topographic region are obtained.

[0009] Furthermore, the formula for calculating the third weight of the pixel is: in, Representing the topographic region in a remote sensing image The internal coordinates are The pixel points and coordinates are The third weight between pixels, Representing the topographic region in a remote sensing image The internal coordinates are Within the eight neighborhoods of a pixel, in the terrain area The coordinates of any pixel within the range; Representing the topographic region in a remote sensing image The internal coordinates are The average pixel value of the pixel in the R, G, and B channels; Representing the topographic region in a remote sensing image The internal coordinates are The average pixel value of the pixel in the R, G, and B channels; Indicates the first smoothing coefficient; This represents the second smoothing coefficient, where the sum of the first smoothing coefficient and the second smoothing coefficient is 1; Representing the topographic region in a remote sensing image The internal coordinates are The gradient value of each pixel.

[0010] Furthermore, the specific method for obtaining the filtering weight of each pixel within the same terrain region based on the third weight of all pixels within the same terrain region and all pixels within the eight-neighborhood of each pixel is as follows: The sum of the third weights of a pixel and all pixels in its eight neighborhoods within the same topographic region is denoted as the weight sum of the pixels within the same topographic region. The result of the calculation of the exponent with the negative of the weight sum as the exponent and the natural constant as the base is denoted as the fourth weight of the pixel. Based on the fourth weight of all pixels within the same terrain area, the filtering weight of each pixel within the same terrain area is obtained.

[0011] Furthermore, the specific method for obtaining the filtering weight of each pixel within the same terrain region based on the fourth weight of all pixels within the same terrain region is as follows: The fourth weight of all pixels in the same terrain area is normalized, and the filtering weight of each pixel in the same terrain area is obtained. The sum of the filtering weights of all pixels in the same terrain area is 1.

[0012] Furthermore, the method for obtaining the corrected illumination component of the terrain area is as follows: The product of the mean pixel value of each pixel in the R, G, and B channels within the terrain region and the filtering weight is denoted as the pixel component of the pixel in the terrain region; the sum of the pixel components of all pixels in the terrain region is denoted as the corrected illumination component of the terrain region.

[0013] Furthermore, the specific method for obtaining the remote sensing enhanced image based on the corrected illumination component of the terrain region in the remote sensing image includes: Using the Retinex image enhancement algorithm, the corrected illumination component of the terrain in high-altitude and cold regions is used as the illumination component of the terrain region to obtain remote sensing enhanced images of high-altitude and cold regions.

[0014] The beneficial effects of this application are: This application addresses the issue that snow accumulation and mountain shadows in high-altitude and cold regions often lead to abnormal reflective points and mountain shadows in remote sensing images, resulting in uneven brightness, low contrast, and loss of regional topographic information. In the process of image enhancement for remote sensing images, the gradient difference between pixels is first used as the texture difference between pixels. Based on the color, spatial location, and texture differences between pixels within the remote sensing image, regions corresponding to different landforms are divided to obtain all landform regions in the remote sensing image. Then, based on the texture, color, and spatial location differences of adjacent pixels within the same landform region, the filtering weight of each pixel within the same landform region is determined. The filtering weight can be applied to the pixels... During image enhancement, the sensitivity to texture differences at pixel locations is adjusted, thereby regulating the smoothness of textures during image enhancement filtering. This reduces pixel edge blurring, preserves terrain details, and further determines the corrected illumination components of the terrain region based on the pixel values ​​and filter weights of all pixels within the terrain area. Finally, based on the corrected illumination components of the terrain region in the remote sensing image, a remote sensing enhanced image is acquired. Using the remote sensing enhanced image, point cloud data, and geographic information system data, an interactive display of a 3D digital model of high-altitude and cold regions is achieved. This addresses the problem of inaccurate 3D digital models in high-altitude and cold regions due to the loss of remote sensing image information, thus improving the quality of 3D digital models in these regions. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1This is a schematic flowchart illustrating an interactive display method for a three-dimensional digital model of a high-altitude and cold region, provided in one embodiment of this application. Figure 2 This is a flowchart illustrating the distance metric acquisition process provided in one embodiment of this application. Detailed Implementation

[0017] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0018] Please see Figure 1 The diagram illustrates a flowchart of an interactive display method for a three-dimensional digital model of a high-altitude, cold region, according to an embodiment of this application. The method includes the following steps: Step S001: Collect remote sensing images, point cloud data, and geographic information system data of high-altitude and cold regions.

[0019] UAVs equipped with optical remote sensing equipment were used to collect remote sensing images of high-altitude and cold regions. These images were RGB images. UAVs also equipped with lidar were used to scan these regions and acquire point cloud data. This point cloud data provides topographic information about the high-altitude and cold regions. Finally, satellite remote sensing technology was used to collect Geographic Information System (GIS) data for these regions.

[0020] Preprocessing is performed on remote sensing images, point cloud data, and geographic information system data from high-altitude and cold regions to improve data quality.

[0021] Preferably, as an embodiment of this application, radiometric and geometric corrections are performed on remote sensing images of high-altitude and cold regions; point cloud data of high-altitude and cold regions are denoised using a statistical outlier removal algorithm; and geographic information system data are subjected to coordinate system unification, attribute cleaning, and topology repair operations to ensure spatial consistency and boundary information closure of geographic information system data, while avoiding duplicate data records.

[0022] Thus, remote sensing images, point cloud data, and geographic information system data for high-altitude and cold regions have been acquired.

[0023] Step S002: Use the SLIC superpixel segmentation algorithm to divide the remote sensing image, obtain the position of all seed points, the color distance and spatial distance between each pixel in the remote sensing image and each seed point, and obtain the metric distance between each pixel in the remote sensing image and each seed point based on the difference in gradient values, color distance and spatial distance between the pixels and seed points in the remote sensing image, as well as the pixel value of each seed point. Based on the metric distance between the pixels and seed points in the remote sensing image, divide the remote sensing image into different landform regions.

[0024] Most areas in high-altitude and cold regions are covered by snow. The surface texture of snow accumulation is not clearly defined, and snow can cause abnormal reflective spots in remote sensing images. Mountains in these regions also affect remote sensing images; areas blocked by mountains appear as shadows, resulting in uneven brightness, low contrast, and loss of regional topographic information. 3D digital models of high-altitude and cold regions built from these images with lost topographic information cannot accurately match the terrain details, thus failing to accurately represent the terrain. Therefore, image enhancement is needed for remote sensing images of high-altitude and cold regions to avoid color distortion caused by strong reflective spots and shadowed areas, thereby improving the quality of 3D digital models of these regions.

[0025] The Retinex image enhancement algorithm can be used to enhance remote sensing images of high-altitude and cold regions by removing low-frequency components, enhancing mid- and high-frequency components, highlighting details and edge information, restoring natural colors, and resolving color distortion caused by strong reflective points and shadow areas. However, while the Retinex algorithm uses the low-pass filtering characteristic of Gaussian filtering to separate the illumination and reflection components and extract low-frequency illumination information, Gaussian filtering indiscriminately smooths noise along with image edges and other details, resulting in noticeable edge blurring and loss of detail in the enhanced image. Therefore, 3D digital models of high-altitude and cold regions built from Retinex-enhanced images cannot accurately represent the terrain and their quality needs improvement.

[0026] To eliminate the impact of snow reflections and mountain shadows on the color quality of remote sensing images and to preserve more topographic details in the enhanced images, it is first necessary to divide the remote sensing images into regions corresponding to different landforms. Then, the illumination and reflection components are extracted based on the texture information of each region to enhance the remote sensing images. This process restores more details while avoiding interference between different regions during the image enhancement process, thus preventing the enhanced remote sensing images from exhibiting obvious edge blurring and loss of detail.

[0027] First, the remote sensing images are divided into regions corresponding to different landforms. In remote sensing images of high-altitude and cold regions, the texture features and color variations of different landform types differ. However, due to the influence of snow cover and mountain shadows in these regions, the distinct landform features within strong reflective areas and shadowed areas are weakened. Furthermore, shadowed areas are often similar to exposed rocks, easily leading to different landforms being grouped into the same region, thus affecting the quality of subsequent remote sensing image enhancement. Therefore, based on the SLIC superpixel segmentation algorithm's region division according to color differences and spatial distances between pixels, texture information differences between pixels are further introduced.

[0028] Convert the remote sensing image to a grayscale image, obtain the remote sensing grayscale image, and calculate the gradient value of all pixels in the remote sensing grayscale image.

[0029] Preferably, this embodiment uses the averaging method to convert the remote sensing image into a grayscale image, and uses the Sobel operator to calculate the gradient values ​​of all pixels in the grayscale remote sensing image. The use of the averaging method to convert the remote sensing image into a grayscale image and the use of the Sobel operator to calculate the gradient values ​​of all pixels in the grayscale remote sensing image are well-known techniques and will not be elaborated further. As other embodiments, based on achieving the purpose of converting the remote sensing image into a grayscale image and calculating the gradient values ​​of all pixels in the grayscale remote sensing image, the implementer may use other methods of the prior art to convert the remote sensing image into a grayscale image and calculate the gradient values ​​of all pixels in the grayscale remote sensing image; this application does not impose any special limitations.

[0030] The remote sensing image was segmented using the SLIC superpixel segmentation algorithm, with the number of superpixels set to [value missing]. , obtain The location of each seed point, the color distance and spatial distance between each pixel in the remote sensing image and each seed point.

[0031] Among them, the use of the SLIC superpixel segmentation algorithm to obtain the position of the seed point, the color distance and spatial distance between the pixel and the seed point are all well-known techniques and will not be described in detail here. This represents the first preset quantity, which is 100 in this embodiment.

[0032] The negative correlation processing result of the mean pixel values ​​of seed points in the R, G, and B channels of the remote sensing image is recorded as the first weight of the seed points in the remote sensing image.

[0033] It is understood that negative correlation processing is applied to the mean pixel values ​​of seed points in the R, G, and B channels of the remote sensing image, ensuring that the mean pixel values ​​of seed points in the R, G, and B channels are negatively correlated with the first weight of the seed points in the remote sensing image. It is understood that the negative correlation in this application refers to the relationship between the independent and dependent variables. The independent variable is the mean pixel values ​​of seed points in the R, G, and B channels of the remote sensing image, and the dependent variable is the first weight of the seed points in the remote sensing image. The negative correlation means that the dependent variable decreases (increases) as the independent variable increases (decreases), and can be an inverse proportional relationship, a subtraction relationship, etc.

[0034] Preferably, as an embodiment of this application, the difference between the ratio of the mean of the pixel values ​​of the seed point in the R, G, and B channels of the remote sensing image to 255 and the number 1 is recorded as the first difference of the seed point in the remote sensing image. The first difference of the seed point in the remote sensing image is used as the result of the calculation of the exponent with the natural constant as the base, and recorded as the first power value of the seed point in the remote sensing image. The difference between the number 1 and the first power value of the seed point in the remote sensing image is recorded as the first weight of the seed point in the remote sensing image.

[0035] Based on the difference in gradient values, color distance, and spatial distance between pixels and seed points in the remote sensing image, as well as the first weight of the seed points, the metric distance between each pixel and each seed point in the remote sensing image is obtained.

[0036] in, Represents pixels in a remote sensing image With seed point The distance between them; Represents seed points in remote sensing images The first weight; Represents pixels in a remote sensing image With seed point Color distance between them; Represents pixels in a remote sensing image With seed point Spatial distance between them; This represents a preset second weight, which is set to 5 in this embodiment; Represents seed points in remote sensing images The number of pixels contained within the superpixel block; Represents pixels in a remote sensing image The gradient value; Represents seed points in remote sensing images The gradient value.

[0037] The smaller the gradient value of a seed point in a remote sensing image, the greater the likelihood that the seed point corresponds to exposed rock or shadow. Therefore, it is crucial to pay close attention to the impact of texture differences between pixels and seed points on region segmentation, avoiding edge blurring and loss of detail caused by similar pixel values. This embodiment uses gradient value differences to evaluate the texture differences between pixels and seed points. The flowchart for distance measurement is shown below. Figure 2 As shown.

[0038] The distance between a pixel and a seed point in the remote sensing image is used as the distance between the pixel and the seed point. The remote sensing image is divided using the SLIC superpixel segmentation algorithm to obtain 100 superpixel regions. All superpixel regions are recorded as landform regions.

[0039] It is understandable that landform features are similar within the same landform region, while landform features differ between different landform regions; a landform region is the result of dividing remote sensing images into regions corresponding to different landforms.

[0040] The use of the SLIC superpixel segmentation algorithm to obtain superpixel regions is a well-known technique and will not be elaborated further.

[0041] This completes the delineation of all landform regions in the remote sensing images of high-altitude and cold regions.

[0042] Step S003: Based on the pixel values, gradient values, and positional differences between pixels within the same topographic region in the remote sensing image and their eight neighboring pixels, calculate the filtering weight for each pixel within the same topographic region. Based on the pixel values ​​and filtering weights of all pixels within the topographic region, determine the corrected illumination component of the topographic region. Based on the corrected illumination component of the topographic region in the remote sensing image, obtain the remote sensing enhanced image.

[0043] After obtaining the segmentation results of different landform regions corresponding to the remote sensing image, the illumination and reflection components of each landform region are extracted based on the texture information of all pixels within that region. Specifically, the illumination component of the landform region is extracted, and then the reflection component is determined based on the illumination component.

[0044] Based on the differences in pixel value, gradient value, and position between pixels within the same geomorphic region and their eight neighboring pixels in the remote sensing image, the third weight of pixels within the same geomorphic region and their eight neighboring pixels is calculated.

[0045] in, Representing the topographic region in a remote sensing image The internal coordinates are The pixel points and coordinates are The third weight between pixels, Representing the topographic region in a remote sensing image The internal coordinates are Within the eight neighborhoods of a pixel, in the terrain area The coordinates of any pixel within the range, Indicates coordinates as The x-coordinate of the pixel. Indicates coordinates as The ordinate value of the pixel. Indicates coordinates as The x-coordinate of the pixel. Indicates coordinates as The ordinate value of the pixel; Representing the topographic region in a remote sensing image The internal coordinates are The average pixel value of the pixel in the R, G, and B channels; Representing the topographic region in a remote sensing image The internal coordinates are The average pixel value of the pixel in the R, G, and B channels; This represents the first smoothing coefficient, which is set to 0.5 in this embodiment. This represents the second smoothing coefficient. In this embodiment, the value of the second smoothing coefficient is 0.5, and the sum of the first smoothing coefficient and the second smoothing coefficient is 1. Representing the topographic region in a remote sensing image The internal coordinates are The gradient value of the pixel; Representing the topographic region in a remote sensing image The internal coordinates are The gradient value of the pixel; Representing the topographic region in a remote sensing image The internal coordinates are The third smoothing coefficient of the pixels; This represents the first parameter, which controls the maximum value of the third smoothing coefficient. In this embodiment, the value of the first parameter is 0.4. This represents the second parameter, and in this embodiment, the value of the second parameter is 5.

[0046] When the gradient value of a pixel within a terrain region in a remote sensing image is larger, the pixel is more likely to be located in a complex, exposed mountain area or a shadowy area with messy textures. In this case, the smaller the third smoothing coefficient of the pixel, the more sensitive it is to texture differences at the pixel location during image enhancement. This makes it easier to reduce the smoothing effect of filtering on texture during image enhancement, thereby minimizing pixel edge blurring and preserving terrain details. Therefore, the third smoothing coefficient of a pixel can improve the enhancement effect of remote sensing images in high-altitude and cold regions. The third weight of a pixel within a terrain region consists of a color term, a spatial term, and a texture term, where the color term is... The spatial term is Texture guide item is .

[0047] The sum of the third weights of a pixel within the same topographic region and all pixels in its eight-neighborhood is denoted as the weight sum of pixels within the same topographic region. The result of calculating the exponent of the weight sum with the negative of the exponent and the natural constant as the base is denoted as the fourth weight of the pixel. The fourth weights of all pixels within the same topographic region are normalized, and the filter weights of each pixel within the same topographic region are obtained separately, so that the sum of the filter weights of all pixels within the same topographic region is 1.

[0048] Preferably, as an embodiment of this application, the sum of the fourth weights of all pixels in the same terrain area is recorded as the first sum of the same terrain area, and the ratio of the fourth weight of the pixel in the terrain area to the first sum of the terrain area where the pixel is located is recorded as the filtering weight of the pixel in the terrain area.

[0049] The corrected illumination component of the terrain region is determined based on the pixel values ​​and filtering weights of all pixels within the terrain region.

[0050] The product of the mean pixel value of each pixel in the R, G, and B channels within the terrain region and the filtering weight is denoted as the pixel component of the pixel in the terrain region; the sum of the pixel components of all pixels in the terrain region is denoted as the corrected illumination component of the terrain region.

[0051] Using the Retinex image enhancement algorithm, the corrected illumination component of the landform area in high-altitude and cold regions is taken as the illumination component of the landform area. Based on the illumination components of all landform areas in the remote sensing image, the remote sensing enhanced image of the high-altitude and cold regions is obtained.

[0052] Among them, the use of the Retinex image enhancement algorithm to enhance remote sensing images is a well-known technique and will not be elaborated further; specifically, the corrected illumination component of the terrain region is used as the illumination component of the terrain region, the reflection component of all terrain regions in the remote sensing image is obtained, and the reflection component is converted to the image domain and nonlinear enhancement is performed using gamma correction to obtain the enhanced remote sensing image of high-altitude and cold regions.

[0053] It is understandable that enhanced remote sensing images of high-altitude and cold regions are simply enhanced remote sensing images of those regions. These enhanced images can provide more clearly defined and textured geomorphic features. Combined with point cloud data and geographic information system (GIS) data from these regions, they can provide accurate geomorphic information for 3D digital modeling of these areas.

[0054] Thus, enhanced remote sensing images of high-altitude and cold regions have been obtained.

[0055] Step S004: Based on remote sensing enhanced images, point cloud data, and geographic information system data, construct a three-dimensional digital model of the high-altitude and cold region, and realize interactive display of the three-dimensional digital model of the high-altitude and cold region.

[0056] Based on enhanced remote sensing images, point cloud data, and geographic information system data of high-altitude and cold regions, a 3D digital model of the region is constructed. This 3D digital model enables interactive display of the region.

[0057] The construction of three-dimensional digital models is a well-known technique and will not be elaborated further.

[0058] This enables interactive display of 3D digital models of high-altitude and cold regions.

[0059] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the principles of this application should be included within the protection scope of this application.

Claims

1. A method for interactively displaying three-dimensional digital models of high-altitude and cold regions, characterized in that, The method includes the following steps: Collect remote sensing images, point cloud data, and geographic information system data from high-altitude and cold regions; The SLIC superpixel segmentation algorithm is used to divide the remote sensing image. The positions of all seed points, the color distance and spatial distance between each pixel and each seed point in the remote sensing image are obtained. Based on the difference in gradient values, color distance and spatial distance between the pixels and seed points in the remote sensing image, as well as the pixel value of each seed point, the metric distance between each pixel and each seed point in the remote sensing image is obtained. Based on the metric distance between the pixels and seed points in the remote sensing image, the remote sensing image is divided into different landform regions. Based on the differences in pixel values, gradient values, and positions between pixels within the same topographic region in the remote sensing image and their eight neighboring pixels, the filtering weight of each pixel within the same topographic region is calculated. Based on the pixel values ​​and filtering weights of all pixels within the topographic region, the corrected illumination component of the topographic region is determined. Based on the corrected illumination component of the topographic region in the remote sensing image, the remote sensing enhanced image is obtained. Based on remote sensing enhanced images, point cloud data, and geographic information system data, a three-dimensional digital model of a high-altitude and cold region is constructed to enable interactive display of the three-dimensional digital model of the region.

2. The interactive display method for a three-dimensional digital model of a high-altitude, cold region according to claim 1, characterized in that, The method for obtaining the metric distance between each pixel and each seed point in the remote sensing image based on the difference in gradient values, color distance, and spatial distance between pixels and seed points in the remote sensing image, as well as the pixel value of each seed point, includes the following specific methods: The negative correlation processing result of the mean pixel values ​​of seed points in the R, G, and B channels of the remote sensing image is recorded as the first weight of the seed points in the remote sensing image. Based on the difference in gradient values, color distance, and spatial distance between pixels and seed points in the remote sensing image, as well as the first weight of the seed points, the metric distance between each pixel and each seed point in the remote sensing image is obtained.

3. The interactive display method for a three-dimensional digital model of a high-altitude, cold region according to claim 2, characterized in that, The formula for calculating the distance between each pixel and each seed point in the remote sensing image is: in, Represents pixels in a remote sensing image With seed point The distance between them; Represents seed points in remote sensing images The first weight; Represents pixels in a remote sensing image With seed point Color distance between them; Represents pixels in a remote sensing image With seed point Spatial distance between them; This indicates the preset second weight; Represents seed points in remote sensing images The number of pixels contained within the superpixel block; Represents pixels in a remote sensing image The gradient value; Represents seed points in remote sensing images The gradient value.

4. The interactive display method for a three-dimensional digital model of a high-altitude, cold region according to claim 1, characterized in that, The method for dividing a remote sensing image into different landform regions based on the metric distance between pixels and seed points in the remote sensing image includes the following specific methods: The distance between a pixel and a seed point in the remote sensing image is used as the distance between the pixel and the seed point. The remote sensing image is divided using the SLIC superpixel segmentation algorithm to obtain superpixel regions, and all superpixel regions are recorded as terrain regions.

5. The interactive display method for a three-dimensional digital model of a high-altitude, cold region according to claim 1, characterized in that, The method for calculating the filtering weight of each pixel within the same topographic region based on the pixel values, gradient values, and positional differences between pixels within the same topographic region and their eight neighboring pixels in the remote sensing image includes the following specific steps: Calculate the topographic region in the remote sensing image The internal coordinates are The third smoothing coefficient for each pixel is calculated as follows: in, Representing the topographic region in a remote sensing image The internal coordinates are The third smoothing coefficient of the pixels; This indicates the preset first parameter; This indicates the preset second parameter; Representing the topographic region in a remote sensing image The internal coordinates are The gradient value of the pixel; Based on the pixel values, gradient values, and positional differences between pixels within the same geomorphic region and pixels within their eight-neighborhood in the remote sensing image, as well as the third smoothing coefficient of the pixels, the third weight of pixels within the same geomorphic region is calculated. Based on the third weights of all pixels within the same topographic region and all pixels in their eight neighborhoods, the filtering weights of each pixel within the same topographic region are obtained.

6. The interactive display method for a three-dimensional digital model of a high-altitude, cold region according to claim 5, characterized in that, The formula for calculating the third weight of the pixel is: in, Representing the topographic region in a remote sensing image The internal coordinates are The pixel points and coordinates are The third weight between pixels, Representing the topographic region in a remote sensing image The internal coordinates are Within the eight neighborhoods of a pixel, in the terrain area The coordinates of any pixel within the range; Representing the topographic region in a remote sensing image The internal coordinates are The average pixel value of the pixel in the R, G, and B channels; Representing the topographic region in a remote sensing image The internal coordinates are The average pixel value of the pixel in the R, G, and B channels; Indicates the first smoothing coefficient; This represents the second smoothing coefficient, where the sum of the first smoothing coefficient and the second smoothing coefficient is 1; Representing the topographic region in a remote sensing image The internal coordinates are The gradient value of each pixel.

7. The interactive display method for a three-dimensional digital model of a high-altitude, cold region according to claim 5, characterized in that, The specific method for obtaining the filtering weight of each pixel within the same topographic region based on the third weight of all pixels within the same topographic region and all pixels within the eight-neighborhood of each pixel is as follows: The sum of the third weights of a pixel and all pixels in its eight neighborhoods within the same topographic region is denoted as the weight sum of the pixels within the same topographic region. The result of the calculation of the exponent with the negative of the weight sum as the exponent and the natural constant as the base is denoted as the fourth weight of the pixel. Based on the fourth weight of all pixels within the same terrain area, the filtering weight of each pixel within the same terrain area is obtained.

8. The interactive display method for a three-dimensional digital model of a high-altitude, cold region according to claim 7, characterized in that, The specific method for obtaining the filtering weight of each pixel within the same topographic region based on the fourth weight of all pixels within the same topographic region is as follows: The fourth weight of all pixels in the same terrain area is normalized, and the filtering weight of each pixel in the same terrain area is obtained. The sum of the filtering weights of all pixels in the same terrain area is 1.

9. The interactive display method for a three-dimensional digital model of a high-altitude, cold region according to claim 1, characterized in that, The method for obtaining the corrected illumination component of the terrain area is as follows: The product of the mean pixel value of each pixel in the R, G, and B channels within the terrain region and the filtering weight is denoted as the pixel component of the pixel in the terrain region; the sum of the pixel components of all pixels in the terrain region is denoted as the corrected illumination component of the terrain region.

10. The interactive display method for a three-dimensional digital model of a high-altitude, cold region according to claim 1, characterized in that, The specific method for obtaining the remote sensing enhanced image based on the corrected illumination component of the terrain region in the remote sensing image includes: Using the Retinex image enhancement algorithm, the corrected illumination component of the terrain in high-altitude and cold regions is used as the illumination component of the terrain region to obtain remote sensing enhanced images of high-altitude and cold regions.

Images