A method, system, device and storage medium for generating a three-dimensional building model

Abstract

The present application provides a method, system, device and storage medium for generating a three-dimensional building model, which collects a set of two-dimensional building images; uses the image texture discrete entropy corresponding to the target building and the given image structure element to convert each two-dimensional building image in the two-dimensional building image set into a two-dimensional morphological image; performs gradient comparison on each two-dimensional morphological image, and then obtains the gradient comparison amount of each two-dimensional morphological image, and maps all two-dimensional morphological images into three-dimensional space based on each gradient comparison amount, and obtains the three-dimensional morphological data of the target building under the three-dimensional gradient; obtains the temperature distribution of the target building, and converts the three-dimensional morphological data into three-dimensional thermal infrared data according to the temperature distribution, and generates a three-dimensional thermal infrared model of the target building through the three-dimensional thermal infrared data. The above scheme can improve the clarity of the texture and detail features in the building image, so as to enhance the realism of the three-dimensional building thermal infrared model.

Description

Technical Field

[0001] The present application relates to the field of image processing technology, and more specifically, to a method, system, device and storage medium for generating a three-dimensional building model. Background Art

[0002] The generation of 3D building models and the image processing involved is a complex process that integrates multidisciplinary technologies such as computer graphics, computer vision, artificial intelligence, and architecture. The generation of 3D building thermal infrared models is a process that combines thermal infrared imaging technology with 3D modeling technology. By collecting thermal infrared data of buildings and mapping them onto 3D models, the thermal performance and energy distribution of buildings can be displayed and analyzed. This process includes the acquisition and processing of thermal infrared images, the fusion of thermal data and geometric data, and the visualization of thermal infrared features on 3D models.

[0003] In the entire process of generating three-dimensional building models, the application of image processing technology runs through. In the modeling stage, geometric modeling technology is usually used to create the three-dimensional shape of the building. The three-dimensional information of the building is usually automatically extracted from the two-dimensional image or video. Through computer vision and deep learning technology, features can be extracted from a large number of building images to generate high-precision three-dimensional building models. However, in the existing technology, when extracting the three-dimensional information of the target building through the building image, due to the low quality of the building image (such as texture mapping distortion, gaps and defects), and the inconsistent relationship between temperature and color mapping, the detailed features in the building thermal infrared model generated based on the building image are blurred, making the building thermal infrared model low in realism and unable to accurately display the thermal performance and energy distribution of the target building. Therefore, how to improve the clarity of the texture and detailed features in the building image to enhance the realism of the three-dimensional building thermal infrared model is a difficult problem faced by the industry. Summary of the invention

[0004] The present application provides a method, system, device and storage medium for generating a three-dimensional building model, which can improve the clarity of texture and detail features in building images to enhance the realism of the three-dimensional building thermal infrared model.

[0005] In a first aspect, the present application provides a method for generating a three-dimensional building model, comprising the following steps:

[0006] Collect two-dimensional images of the target building at different angles to obtain a two-dimensional image set of the building;

[0007] Determine the texture entropy of each two-dimensional building image in the two-dimensional building image set, use all the texture entropies to determine the image texture discrete entropy corresponding to the target building, and convert each two-dimensional building image in the two-dimensional building image set into a two-dimensional morphological image according to the image texture discrete entropy and a given image structure element;

[0008] Perform gradient comparison on each two-dimensional morphological image respectively, and then obtain the gradient comparison amount of each two-dimensional morphological image, and map all the two-dimensional morphological images into three-dimensional space based on each gradient comparison amount, and obtain the stereoscopic morphological data of the target building under the three-dimensional gradient;

[0009] The temperature distribution of the target building is acquired, the three-dimensional morphological data is converted into three-dimensional thermal infrared data according to the temperature distribution, and a three-dimensional thermal infrared model of the target building is generated through the three-dimensional thermal infrared data.

[0010] In some embodiments, before determining the texture entropy of each two-dimensional building image in the two-dimensional building image set, the method further includes: performing grayscale inversion on each two-dimensional building image in the two-dimensional building image set.

[0011] In some embodiments, performing gradient comparison on each two-dimensional morphological image respectively, and then obtaining the gradient comparison amount of each two-dimensional morphological image specifically includes:

[0012] For each two-dimensional morphological image, gradient comparison is performed on each pixel point in the two-dimensional morphological image, thereby obtaining a gradient comparison value corresponding to each pixel point in the two-dimensional morphological image;

[0013] The gradient comparison amount of the two-dimensional morphological image is determined by all the gradient comparison values, and then the gradient comparison amount of each two-dimensional morphological image is obtained.

[0014] In some embodiments, mapping all two-dimensional morphological images into three-dimensional space based on each gradient comparison amount to obtain stereoscopic morphological data of the target building under three-dimensional gradient specifically includes:

[0015] All two-dimensional morphological images are screened according to each gradient comparison amount, and then the screened two-dimensional morphological images are mapped into three-dimensional space according to the corresponding gradient comparison amount to obtain the volume data of the target building under the three-dimensional gradient;

[0016] The volume data of the target building under the three-dimensional gradient is grid-divided to obtain the stereoscopic morphological data of the target building under the three-dimensional gradient.

[0017] In some embodiments, a thermal infrared imaging device is used to obtain the temperature distribution of the target building.

[0018] In some embodiments, converting the stereoscopic morphological data into three-dimensional thermal infrared data according to the temperature distribution specifically includes:

[0019] Determine the temperature-color mapping domain corresponding to the target building;

[0020] determining a temperature color distribution of a target building based on the temperature distribution and the temperature-color mapping domain;

[0021] The temperature color attribute is assigned to the three-dimensional morphological data through the temperature color distribution to obtain the three-dimensional thermal infrared data of the target building.

[0022] In some embodiments, determining the temperature-color mapping domain corresponding to the target building specifically includes:

[0023] Obtain the temperature distribution range of the area where the target building is located;

[0024] Eliminating extreme values ​​in the temperature distribution interval, and then determining the temperature mapping domain of the target building according to the temperature distribution interval after eliminating the extreme values;

[0025] The temperature mapping domain is divided according to preset temperature intervals, and then the divided temperature mapping domain is converted into a temperature-color mapping domain corresponding to the target building using a color model.

[0026] In a second aspect, the present application provides a system for generating a three-dimensional building model, comprising:

[0027] The acquisition module is used to acquire two-dimensional images of the target building at different angles to obtain a set of two-dimensional images of the building;

[0028] A processing module is used to determine the texture entropy of each two-dimensional building image in the two-dimensional building image set, use all the texture entropies to determine the image texture discrete entropy corresponding to the target building, and convert each two-dimensional building image in the two-dimensional building image set into a two-dimensional morphological image according to the image texture discrete entropy and a given image structure element;

[0029] The processing module is further used to perform gradient comparison on each two-dimensional morphological image respectively, thereby obtaining a gradient comparison amount of each two-dimensional morphological image, and mapping all two-dimensional morphological images into three-dimensional space based on each gradient comparison amount to obtain stereoscopic morphological data of the target building under the three-dimensional gradient;

[0030] The execution module is used to obtain the temperature distribution of the target building, convert the three-dimensional morphological data into three-dimensional thermal infrared data according to the temperature distribution, and generate a three-dimensional thermal infrared model of the target building through the three-dimensional thermal infrared data.

[0031] In a third aspect, the present application provides a computer device, comprising a memory and a processor, wherein the memory is used to store a computer program, and the processor is used to call and run the computer program from the memory, so that the computer device executes the above-mentioned method for generating a three-dimensional building model.

[0032] In a fourth aspect, the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores instructions or codes, and when the instructions or codes are run on a computer, the computer implements the above-mentioned method for generating a three-dimensional building model when executing the computer.

[0033] The technical solution provided by the embodiments disclosed in this application has the following beneficial effects:

[0034] The present application provides a method, system, device and storage medium for generating a three-dimensional building model. The method comprises the following steps: collecting two-dimensional images of a target building at different angles to obtain a set of two-dimensional building images; determining the texture entropy of each two-dimensional building image in the set of two-dimensional building images, using all the texture entropies to determine the image texture discrete entropy corresponding to the target building, and converting each two-dimensional building image in the set of two-dimensional building images into a two-dimensional morphological image according to the image texture discrete entropy and a given image structure element; performing gradient comparison on each two-dimensional morphological image to obtain a gradient comparison amount of each two-dimensional morphological image, mapping all the two-dimensional morphological images into a three-dimensional space based on each gradient comparison amount, and obtaining stereoscopic morphological data of the target building under a three-dimensional gradient; obtaining the temperature distribution of the target building, converting the stereoscopic morphological data into three-dimensional thermal infrared data according to the temperature distribution, and generating a three-dimensional thermal infrared model of the target building through the three-dimensional thermal infrared data.

[0035] It can be seen that in the present application, the clarity of the texture and detail features in the building image can be improved to enhance the realism of the three-dimensional building thermal infrared model; wherein, by calculating the discrete entropy of the image texture, the detail features in the building image can be quantified and highlighted, and the detail enhancement helps to more accurately represent the various materials and structural features on the surface of the building, and morphological processing (such as expansion, corrosion, etc.) is performed according to the discrete entropy of the image texture and the image structure elements, so that important structural features can be retained and enhanced; then, the difference in the gradient of the texture features of the two-dimensional morphological image can be quantified by gradient comparison and gradient comparison amount calculation, so that the two-dimensional morphological image used for three-dimensional reconstruction can be ensured to have clearer texture details, and the three-dimensional morphological data of the building can be generated more accurately based on the three-dimensional mapping of the gradient comparison amount; finally, using the temperature distribution of the target building, it can be ensured that the three-dimensional building thermal infrared model reflects the temperature distribution of the target building in the actual environment, and by converting the three-dimensional morphological data into three-dimensional thermal infrared data, the temperature change can be made more obvious, the temperature contrast in the three-dimensional building thermal infrared model can be enhanced, and the texture and detail features in the three-dimensional building thermal infrared model can be highlighted. In summary, the technical solution adopted in this application can improve the clarity of texture and detail features in building images to enhance the realism of three-dimensional building thermal infrared models. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. 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 creative labor.

[0037] Figure 1 is an exemplary flow chart of a method for generating a three-dimensional building model according to some embodiments of the present application;

[0038] Figure 2 is an exemplary flow chart of determining a two-dimensional morphological image according to some embodiments of the present application;

[0039] Figure 3 is an exemplary flow chart for determining three-dimensional thermal infrared data according to some embodiments of the present application;

[0040] Figure 4 is a schematic diagram of exemplary hardware and / or software of a system for generating a three-dimensional building model according to some embodiments of the present application;

[0041] Figure 5 It is a structural schematic diagram of a computer device for implementing a method for generating a three-dimensional building model according to some embodiments of the present application. DETAILED DESCRIPTION

[0042] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0043] The embodiment of the present application provides a method, system, device and storage medium for generating a three-dimensional building model, the core of which is to collect two-dimensional images of target buildings at different angles to obtain a set of two-dimensional building images; determine the texture entropy of each two-dimensional building image in the set of two-dimensional building images, use all texture entropies to determine the image texture discrete entropy corresponding to the target building, and convert each two-dimensional building image in the set of two-dimensional building images into a two-dimensional morphological image according to the image texture discrete entropy and a given image structure element; perform gradient comparison on each two-dimensional morphological image respectively, and then obtain the gradient comparison amount of each two-dimensional morphological image, and map all two-dimensional morphological images into three-dimensional space based on each gradient comparison amount to obtain the three-dimensional morphological data of the target building under the three-dimensional gradient; obtain the temperature distribution of the target building, convert the three-dimensional morphological data into three-dimensional thermal infrared data according to the temperature distribution, and generate a three-dimensional thermal infrared model of the target building through the three-dimensional thermal infrared data. The above scheme can improve the clarity of texture and detail features in the building image to enhance the realism of the three-dimensional building thermal infrared model.

[0044] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods. Figure 1 , which is an exemplary flow chart of a method for generating a three-dimensional building model according to some embodiments of the present application. The method 100 for generating a three-dimensional building model mainly includes the following steps:

[0045] In step 101, two-dimensional images of a target building at different angles are collected to obtain a set of two-dimensional images of the building.

[0046] In specific implementation, two-dimensional images of the target building can be taken from different angles by a drone, so as to obtain multiple two-dimensional images of the building. The collection of all the two-dimensional images of the building can be used as a two-dimensional image set of the building. It should be noted that, according to the scale and shape of the target building, multiple shooting positions are selected to ensure that all sides of the target building are covered, including the front, side, back and top. At each shooting position, the camera angle is adjusted to ensure that the target building can be photographed from various heights and oblique angles. Through the combination of vertical, oblique and horizontal angles, more structural details of the target building can be obtained.

[0047] In step 102, the texture entropy of each building two-dimensional image in the building two-dimensional image set is determined, and the image texture discrete entropy corresponding to the target building is determined using all the texture entropies. According to the image texture discrete entropy and a given image structure element, each building two-dimensional image in the building two-dimensional image set is converted into a two-dimensional morphological image.

[0048] In some embodiments, before determining the texture entropy of each building two-dimensional image in the building two-dimensional image set, the method also includes: performing grayscale inversion on each building two-dimensional image in the building two-dimensional image set; in specific implementation, for each building two-dimensional image in the building two-dimensional image set, the building two-dimensional image can be converted into a grayscale image, and the grayscale image can be grayscale inverted, that is, each grayscale value in the grayscale image is subtracted from 255. In the above manner, each building two-dimensional image in the building two-dimensional image set can be grayscale inverted.

[0049] In some embodiments, the texture entropy of each building two-dimensional image in the building two-dimensional image set is determined; in specific implementation, for each building two-dimensional image in the building two-dimensional image set, the entropy value of the building two-dimensional image can be calculated based on Shannon theory, and the product of the entropy value of the building two-dimensional image and a preset proportional coefficient is used as the texture entropy of the building two-dimensional image. The texture entropy of each building two-dimensional image in the building two-dimensional image set can be determined in the above manner. In the present application, the value of the proportional coefficient is 0.78. In actual implementation, the proportional coefficient can also take other values, which are not limited here. It should be noted that in the present application, texture entropy represents the complexity of texture features in the building two-dimensional image. The larger the texture entropy, the richer the detail features of the building two-dimensional image and the better the image quality.

[0050] Preferably, in some embodiments, all texture entropies are used to determine the image texture discrete entropy corresponding to the target building; in specific implementation, the image texture discrete entropy corresponding to the target building can be determined by all texture entropies, that is, the standard deviation of all texture entropies can be used as the image texture discrete entropy corresponding to the target building. In other embodiments, the average value of all texture entropies can also be used as the image texture discrete entropy corresponding to the target building, which will not be repeated here; it should be noted that, in the present application, the image texture discrete entropy represents the average complexity of the detail texture features of all two-dimensional images of the target building.

[0051] Preferably, in some embodiments, reference Figure 2 As shown, this figure is an exemplary flow chart of determining a two-dimensional morphological image in some embodiments of the present application. In this embodiment, converting each two-dimensional building image in the two-dimensional building image set into a two-dimensional morphological image according to the image texture discrete entropy and the given image structure element can be implemented by the following steps:

[0052] First, in step 1021, for each two-dimensional building image in the two-dimensional building image set, the two-dimensional building image is corroded according to the image texture discrete entropy and a given image structure element to obtain a corroded image corresponding to the two-dimensional building image;

[0053] Then, in step 1022, the eroded image is expanded according to the image texture discrete entropy and a given image structure element to obtain an expanded image corresponding to the two-dimensional image of the building;

[0054] Finally, in step 1023, the two-dimensional building image is converted into a two-dimensional morphological image through the erosion image, the dilation image and the image structure element, and then each two-dimensional building image in the two-dimensional building image set is converted into a two-dimensional morphological image.

[0055] It should be noted that in the present application, a given image structure element is a predefined rectangle used to detect structural features in an image. The image structure element is usually a two-dimensional matrix, which contains pixel values, usually 0 and 1, and the size of the image structure element can be determined by historical experience. In specific implementation, first, for each two-dimensional building image in the building two-dimensional image set, the building two-dimensional image can be eroded according to the image texture discrete entropy and the given image structure element. For example, the image structure element is set to L(x, y), the building two-dimensional image is H(j, k), and the image texture discrete entropy is a. Then, the eroded image F(j, k) corresponding to the building two-dimensional image can be expressed as: F(j, k)=max[H(jx, ky)]+a*L(x, y), where (x, y)∈b, b represents the definition domain corresponding to the building two-dimensional image. The eroded image corresponding to the building two-dimensional image can be obtained in the above manner. Then, the image texture can be used to obtain the eroded image F(j, k) corresponding to the building two-dimensional image. The eroded image is expanded by using the discrete entropy and the given image structure element. For example, the image structure element is set to be L(x, y), the eroded image is set to be F(j, k), and the image texture discrete entropy is set to be a. Then the expanded image P(j, k) corresponding to the two-dimensional image of the building can be expressed as: P(j, k)=min[F(jx, ky)]-a*L(x, y), wherein (x, y)∈b, b represents the definition domain corresponding to the two-dimensional image of the building. The expanded image corresponding to the two-dimensional image of the building can be obtained by the above method. Finally, the two-dimensional image of the building is converted into a two-dimensional morphological image by using the eroded image, the expanded image and the image structure element. The product of the three images, the eroded image, the expanded image and the image structure element is used as the two-dimensional morphological image corresponding to the two-dimensional image of the building. The two-dimensional morphological image is a two-dimensional image that clearly represents the morphological structural features and details of the target building. The two-dimensional image of the building in the two-dimensional image set of the building can be converted into a two-dimensional morphological image by the above method.

[0056] It should be noted that through the calculation of image texture discrete entropy, the detail features in the building image can be quantified and highlighted, and detail enhancement helps to more accurately represent the various materials and structural features on the building surface. Morphological processing (such as expansion, corrosion, etc.) based on image texture discrete entropy and image structural elements can effectively remove noise while retaining and enhancing important structural features. The realism of the three-dimensional building thermal infrared model depends on the detail of the surface texture. By enhancing the image details through image texture discrete entropy, the texture in the thermal infrared image can be made more realistic.

[0057] In step 103, gradient comparison is performed on each two-dimensional morphological image to obtain the gradient comparison value of each two-dimensional morphological image. Based on each gradient comparison value, all two-dimensional morphological images are mapped into three-dimensional space to obtain stereoscopic morphological data of the target building under three-dimensional gradient.

[0058] In some embodiments, the gradient comparison is performed on each two-dimensional morphological image respectively, and then the gradient comparison amount of each two-dimensional morphological image is obtained in the following manner, namely:

[0059] For each two-dimensional morphological image, gradient comparison is performed on each pixel point in the two-dimensional morphological image, thereby obtaining a gradient comparison value corresponding to each pixel point in the two-dimensional morphological image;

[0060] The gradient comparison amount of the two-dimensional morphological image is determined by all the gradient comparison values, and then the gradient comparison amount of each two-dimensional morphological image is obtained.

[0061] It should be noted that, in the present application, the gradient comparison amount is an indicator used to measure the gradient difference of the texture features of the two-dimensional morphological image. The larger the gradient comparison amount, the higher the gradient difference between the texture features in the two-dimensional morphological image, and the richer the texture features contained in the two-dimensional morphological image; the gradient comparison value refers to the gradient difference between the pixels in the two-dimensional morphological image; in specific implementation, first, for each two-dimensional morphological image, a gradient comparison is performed on each pixel in the two-dimensional morphological image, for example, given a pixel I (x, y) in a two-dimensional morphological image I, the gradient comparison value corresponding to the pixel I (x, y) can be obtained by the following method: The difference between the pixel point I(x+2, y) and the pixel point I(x, y) is calculated, and the square of the difference is used as the gradient comparison value corresponding to the pixel point I(x, y). The gradient comparison value corresponding to each pixel point in the two-dimensional morphological image can be obtained in the above manner; then, the gradient comparison amount of the two-dimensional morphological image is determined by all the gradient comparison values, and the sum of all the gradient comparison values ​​can be used as the gradient comparison amount of the two-dimensional morphological image. In other embodiments, other methods can also be used to determine the gradient comparison amount of the two-dimensional morphological image, which will not be repeated here. The gradient comparison amount of each two-dimensional morphological image can be obtained in the above manner.

[0062] In some embodiments, all two-dimensional morphological images are mapped into three-dimensional space based on various gradient comparison amounts, and the stereoscopic morphological data of the target building under the three-dimensional gradient can be obtained in the following manner, namely:

[0063] All two-dimensional morphological images are screened according to each gradient comparison amount, and then the screened two-dimensional morphological images are mapped into three-dimensional space according to the corresponding gradient comparison amount to obtain the volume data of the target building under the three-dimensional gradient;

[0064] The volume data of the target building under the three-dimensional gradient is grid-divided to obtain the stereoscopic morphological data of the target building under the three-dimensional gradient.

[0065] In the specific implementation, first, all the two-dimensional morphological images are screened according to each gradient comparison amount, that is, the two-dimensional morphological images with gradient comparison amounts greater than a set threshold are screened out. The threshold can be set through historical experiments and is not limited here; then, for each screened two-dimensional morphological image, the pixel points in the two-dimensional morphological image are mapped to the three-dimensional space according to the shooting position and angle of each two-dimensional morphological image. For the overlapping area between the two-dimensional morphological images, the gradient comparison amount corresponding to the two-dimensional morphological image can be used as a weight for mapping, and the triangulation algorithm can be used to map the pixel points of the two-dimensional morphological image to the three-dimensional space to generate the target building in the three-dimensional space. The volume data under the gradient, the triangulation algorithm uses the overlapping area between the two-dimensional morphological images and the camera position to calculate the corresponding three-dimensional coordinate points; finally, the generated volume data is gridded. The gridding can use space segmentation algorithms such as Octree or Delaunay triangulation. The purpose of gridding is to organize the volume data of the target building under the three-dimensional gradient into structured data for subsequent processing and analysis. The volume data of the target building under the three-dimensional gradient after gridding is the three-dimensional morphological data of the target building, which is data representing the three-dimensional morphology and detail features of the target building.

[0066] It should be noted that the texture details and complexity of the two-dimensional morphological image can be quantified through gradient evaluation and gradient comparison calculation, thereby ensuring that the two-dimensional morphological image used for three-dimensional reconstruction has high details and high texture entropy, and the three-dimensional mapping based on the gradient comparison can more accurately generate three-dimensional point cloud data of the exterior of the building, ensuring the accuracy and integrity of the point cloud data.

[0067] In step 104, the temperature distribution of the target building is obtained, the three-dimensional morphological data is converted into three-dimensional thermal infrared data according to the temperature distribution, and a three-dimensional thermal infrared model of the target building is generated through the three-dimensional thermal infrared data.

[0068] In some embodiments, a thermal infrared imaging device can be used to obtain the temperature distribution of the target building. In specific implementation, a thermal infrared imaging device (such as an infrared camera, a thermal imager) can be used to photograph the target building from multiple angles to obtain the temperature distribution of the target building, that is, the temperature of each point of the target building. In some embodiments, the collected temperature distribution data can be processed and calibrated to ensure the accuracy and consistency of the data. The processing steps may include noise filtering, data smoothing, calibration and registration, etc.

[0069] Preferably, in some embodiments, reference Figure 3 As shown, this figure is an exemplary flow chart for determining three-dimensional thermal infrared data in some embodiments of the present application. In this embodiment, converting the three-dimensional morphological data into three-dimensional thermal infrared data according to the temperature distribution can be achieved by the following steps:

[0070] First, in step 1041, the temperature-color mapping domain corresponding to the target building is determined;

[0071] Then, in step 1042, a temperature color distribution of the target building is determined based on the temperature distribution and the temperature-color mapping domain;

[0072] Finally, in step 1043, temperature color attributes are assigned to the three-dimensional morphological data through the temperature color distribution to obtain three-dimensional thermal infrared data of the target building.

[0073] In specific implementation, first, the temperature-color mapping domain corresponding to the target building can be determined, and the temperature-color mapping domain represents the mapping interval of each temperature of the target building and the color corresponding to the temperature; then, the temperature value of each location of the target building can be obtained according to the temperature distribution of the target building, so that the temperature distribution can be matched with the temperature-color mapping domain to obtain the temperature color distribution of the target building, that is, the color that should be assigned to each point of the target building; then, the color of each location in the temperature color distribution can be assigned as the temperature color attribute of the corresponding point in the three-dimensional morphological data, so as to obtain the three-dimensional thermal infrared data of the target building, and the three-dimensional thermal infrared data contains the spatial coordinates of each point and its corresponding color (reflecting the temperature information).

[0074] In the above embodiment, the temperature-color mapping domain corresponding to the target building may be determined in the following manner, namely:

[0075] Obtain the temperature distribution range of the area where the target building is located;

[0076] Eliminating extreme values ​​in the temperature distribution interval, and then determining the temperature mapping domain of the target building according to the temperature distribution interval after eliminating the extreme values;

[0077] The temperature mapping domain is divided according to preset temperature intervals, and then the divided temperature mapping domain is converted into a temperature-color mapping domain corresponding to the target building using a color model.

[0078] In specific implementation, first, the temperature distribution range of the target building can be obtained from the local meteorological bureau or the National Meteorological Data Center, or through satellite data or meteorological websites. For example, the temperature distribution range of the target building can be [-12°C, 6°C]; then, possible extreme values ​​can be identified and eliminated. In this application, extreme values ​​refer to temperature data points that significantly deviate from the overall trend, which may be caused by instrument errors or short-term weather changes. Statistical methods such as box plots can be used to detect and eliminate extreme values, or values ​​in the temperature data that exceed the mean ±3 standard deviations can be eliminated based on empirical rules. After eliminating extreme values, the temperature distribution range can be recalculated. For example, after eliminating extreme values, the new temperature distribution range may be [-10°C, 5°C]; secondly, in order to cover more actual situations and changes in building surface temperature, the temperature distribution range can be appropriately expanded. For example, the minimum temperature and the maximum temperature are respectively expanded by a certain range (such as 2-5°C). For example, the temperature distribution range [-10°C, 5°C] is expanded to [-15°C, 10°C], the expanded temperature distribution range is the temperature mapping domain, which is used for subsequent temperature-to-color mapping to show a wider range of temperature changes in the thermal infrared model.

[0079] In addition, a suitable temperature interval can be preset to divide the entire temperature mapping domain into several sub-intervals. This interval can be set according to demand, such as one interval every 2°C or every 1°C; then, the temperature mapping domain can be divided into several sub-intervals according to the preset temperature interval. For example, if the preset temperature interval is 2°C, the temperature mapping domain [-15°C, 10°C] can be divided into multiple sub-intervals; finally, using a color model to uniformly stretch the divided temperature mapping domain refers to selecting a suitable color model for mapping temperature to color. Commonly used color models include HSV (hue-saturation-lightness), RGB (red-green-blue), etc., and assigning a corresponding color range to each sub-interval of the divided temperature mapping domain. For example, the lowest temperature can be mapped to blue (hue 0) and the highest temperature can be mapped to red (hue 1) in the HSV hue space, and the intermediate temperatures are uniformly distributed in turn, that is, each divided temperature mapping domain is uniformly stretched and mapped to the corresponding range of the color model, so that the color range mapped to each sub-interval in the divided temperature mapping domain is the same, and the divided temperature mapping domain is uniformly stretched, thereby obtaining the temperature-color mapping domain corresponding to the target building, and the temperature-color mapping domain indicates that each temperature value of the target building corresponds to a color.

[0080] It should be noted that using the temperature distribution of the target building can ensure that the three-dimensional building thermal infrared model reflects the temperature distribution of the target building in the actual environment. The temperature mapping domain is stretched using the color model, and the temperature distribution is mapped into the color range to make the temperature change more obvious. This can enhance the temperature contrast in the three-dimensional building thermal infrared model and highlight the texture and detail features in the three-dimensional building thermal infrared model.

[0081] In some embodiments, a three-dimensional thermal infrared model of the target building is generated using the three-dimensional thermal infrared data. In specific implementation, computer graphics technology can be used to visualize the three-dimensional thermal infrared data to generate a three-dimensional thermal infrared model. The model can be optimized using triangulated meshing, surface reconstruction and other methods to make its structure clearer and its details more realistic.

[0082] It can be seen that in the present application, the clarity of texture and detail features in the building image can be improved to enhance the realism of the three-dimensional building thermal infrared model; wherein, by calculating the discrete entropy of the image texture, the detail features in the building image can be quantified and highlighted, and the detail enhancement helps to more accurately represent the various materials and structural features on the surface of the building, and morphological processing (such as expansion, corrosion, etc.) is performed according to the discrete entropy of the image texture and the image structure elements, so that important structural features can be retained and enhanced; then, the difference in the gradient of the texture features of the two-dimensional morphological image can be quantified by gradient comparison and gradient comparison amount calculation, so that the two-dimensional morphological image used for three-dimensional reconstruction can be ensured to have high details, and the three-dimensional morphological data of the building can be generated more accurately based on the three-dimensional mapping of the gradient comparison amount; finally, using the temperature distribution of the target building, it can be ensured that the three-dimensional building thermal infrared model reflects the temperature distribution of the target building in the actual environment, and by converting the three-dimensional morphological data into three-dimensional thermal infrared data, the temperature change can be made more obvious, the temperature contrast in the three-dimensional building thermal infrared model can be enhanced, and the texture and detail features in the three-dimensional building thermal infrared model can be highlighted. In summary, the technical solution adopted in this application can improve the clarity of texture and detail features in building images to enhance the realism of three-dimensional building thermal infrared models.

[0083] In addition, in another aspect of the present application, in some embodiments, the present application provides a system for generating a three-dimensional building model, referring to Figure 4 , which is a schematic diagram of exemplary hardware and / or software of a system for generating a three-dimensional building model according to some embodiments of the present application, wherein the system 400 for generating a three-dimensional building model comprises: a collection module 401, a processing module 402 and an execution module 403, which are described as follows:

[0084] The acquisition module 401 in the present application is mainly used to acquire two-dimensional images of the target building at different angles to obtain a set of two-dimensional images of the building;

[0085] Processing module 402, in the present application, the processing module 402 is used to determine the texture entropy of each building two-dimensional image in the building two-dimensional image set, use all the texture entropies to determine the image texture discrete entropy corresponding to the target building, and convert each building two-dimensional image in the building two-dimensional image set into a two-dimensional morphological image according to the image texture discrete entropy and a given image structure element;

[0086] It should be noted that the processing module 402 in the present application is also used to perform gradient comparison on each two-dimensional morphological image, thereby obtaining a gradient comparison amount of each two-dimensional morphological image, and mapping all two-dimensional morphological images into three-dimensional space based on each gradient comparison amount to obtain stereoscopic morphological data of the target building under the three-dimensional gradient;

[0087] Execution module 403, in this application, execution module 403 is mainly used to obtain the temperature distribution of the target building, convert the three-dimensional morphological data into three-dimensional thermal infrared data according to the temperature distribution, and generate a three-dimensional thermal infrared model of the target building through the three-dimensional thermal infrared data.

[0088] The above describes in detail the examples of the generation method, system, device and storage medium of the three-dimensional building model provided in the embodiments of the present application. It can be understood that in order to realize the above functions, the corresponding device includes the hardware structure and / or software module corresponding to the execution of each function. It should be easily appreciated by those skilled in the art that, in combination with the units and algorithm steps of each example described in the embodiments disclosed herein, the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is executed in the form of hardware or computer software driving hardware depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present application.

[0089] In some embodiments, the present application also provides a computer device, comprising a memory and a processor, wherein the memory is used to store a computer program, and the processor is used to call and run the computer program from the memory, so that the computer device executes the above-mentioned method for generating a three-dimensional building model.

[0090] In some embodiments, reference Figure 5, the dotted line in the figure indicates that the unit or the module is optional, and the figure is a schematic diagram of the structure of a computer device for implementing a method for generating a three-dimensional building model according to an embodiment of the present application. The method for generating a three-dimensional building model in the above embodiment can be Figure 5 The computer device 500 shown in the figure is implemented, and the computer device 500 includes at least one processor 501, a memory 502 and at least one communication unit 505. The computer device 500 can be a terminal device, a server or a chip.

[0091] The processor 501 may be a general-purpose processor or a special-purpose processor. For example, the processor 501 may be a central processing unit (CPU), which may be used to control the computer device 500, execute software programs, and process data of the software programs. The computer device 500 may also include a communication unit 505 to implement signal input (reception) and output (transmission).

[0092] For example, the computer device 500 may be a chip, the communication unit 505 may be an input and / or output circuit of the chip, or the communication unit 505 may be a communication interface of the chip, and the chip may be a component of a terminal device, a network device, or other devices.

[0093] For another example, the computer device 500 may be a terminal device or a server, and the communication unit 505 may be a transceiver of the terminal device or the server, or the communication unit 505 may be a transceiver circuit of the terminal device or the server.

[0094] The computer device 500 may include one or more memories 502, on which a program 504 is stored. The program 504 can be executed by the processor 501 to generate instructions 503, so that the processor 501 performs the method described in the above method embodiment according to the instructions 503. Optionally, data (such as a target audit model) can also be stored in the memory 502. Optionally, the processor 501 can also read the data stored in the memory 502, and the data can be stored at the same storage address as the program 504, or the data can be stored at a different storage address from the program 504.

[0095] The processor 501 and the memory 502 may be provided separately or integrated together, for example, integrated on a system on chip (SOC) of the terminal device.

[0096] It should be understood that each step of the above method embodiment can be completed by a hardware-based logic circuit or software-based instructions in the processor 501. The processor 501 can be a central processing unit, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, such as discrete gates, transistor logic devices or discrete hardware components.

[0097] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0098] For example, in some embodiments, the present application also provides a computer-readable storage medium, in which instructions or codes are stored. When the instructions or codes are run on a computer, the computer implements the above-mentioned method for generating a three-dimensional building model when executing the computer.

[0099] Although the preferred embodiments of the present application have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present application.

[0100] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and 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 also intended to include these modifications and variations.

Claims

1. A method for generating a three-dimensional building model, characterized in that: The steps include: Collect two-dimensional images of the target building at different angles to obtain a two-dimensional image set of the building; Determine the texture entropy of each two-dimensional building image in the two-dimensional building image set, use all the texture entropies to determine the image texture discrete entropy corresponding to the target building, and convert each two-dimensional building image in the two-dimensional building image set into a two-dimensional morphological image according to the image texture discrete entropy and a given image structure element; Perform gradient comparison on each two-dimensional morphological image respectively, and then obtain the gradient comparison amount of each two-dimensional morphological image, and map all the two-dimensional morphological images into three-dimensional space based on each gradient comparison amount, and obtain the stereoscopic morphological data of the target building under the three-dimensional gradient; Acquire the temperature distribution of the target building, convert the three-dimensional morphological data into three-dimensional thermal infrared data according to the temperature distribution, and generate a three-dimensional thermal infrared model of the target building through the three-dimensional thermal infrared data; The gradient comparison is performed on each two-dimensional morphological image respectively, and the gradient comparison amount of each two-dimensional morphological image is obtained, which specifically includes: For each two-dimensional morphological image, gradient comparison is performed on each pixel point in the two-dimensional morphological image, thereby obtaining a gradient comparison value corresponding to each pixel point in the two-dimensional morphological image; Determine the gradient comparison amount of the two-dimensional morphological image through all the gradient comparison values, and then obtain the gradient comparison amount of each two-dimensional morphological image; Among them, based on each gradient comparison amount, all two-dimensional morphological images are mapped into three-dimensional space, and the stereoscopic morphological data of the target building under the three-dimensional gradient is obtained, which specifically includes: All two-dimensional morphological images are screened according to each gradient comparison amount, and then the screened two-dimensional morphological images are mapped into three-dimensional space according to the corresponding gradient comparison amount to obtain the volume data of the target building under the three-dimensional gradient; The volume data of the target building under the three-dimensional gradient is grid-divided to obtain the stereoscopic morphological data of the target building under the three-dimensional gradient.

2. The method according to claim 1, characterized in that Before determining the texture entropy of each two-dimensional building image in the two-dimensional building image set, the method further includes: performing grayscale inversion on each two-dimensional building image in the two-dimensional building image set.

3. The method according to claim 1, characterized in that Use thermal infrared imaging equipment to obtain the temperature distribution of the target building.

4. The method according to claim 1, characterized in that Converting the three-dimensional morphological data into three-dimensional thermal infrared data according to the temperature distribution specifically includes: Determine the temperature-color mapping domain corresponding to the target building; determining a temperature color distribution of a target building based on the temperature distribution and the temperature-color mapping domain; The temperature color attribute is assigned to the three-dimensional morphological data through the temperature color distribution to obtain the three-dimensional thermal infrared data of the target building.

5. The method according to claim 4, characterized in that Determine the temperature-color mapping domain corresponding to the target building specifically includes: Obtain the temperature distribution range of the area where the target building is located; Eliminating extreme values ​​in the temperature distribution interval, and then determining the temperature mapping domain of the target building according to the temperature distribution interval after eliminating the extreme values; The temperature mapping domain is divided according to preset temperature intervals, and then the divided temperature mapping domain is converted into a temperature-color mapping domain corresponding to the target building using a color model.

6. A system for generating a three-dimensional building model, which generates a three-dimensional building model using the method according to any one of claims 1 to 5, characterized in that: The generation system of the three-dimensional building model includes: The acquisition module is used to acquire two-dimensional images of the target building at different angles to obtain a set of two-dimensional images of the building; A processing module is used to determine the texture entropy of each two-dimensional building image in the two-dimensional building image set, use all the texture entropies to determine the image texture discrete entropy corresponding to the target building, and convert each two-dimensional building image in the two-dimensional building image set into a two-dimensional morphological image according to the image texture discrete entropy and a given image structure element; The processing module is further used to perform gradient comparison on each two-dimensional morphological image respectively, thereby obtaining a gradient comparison amount of each two-dimensional morphological image, and mapping all two-dimensional morphological images into three-dimensional space based on each gradient comparison amount to obtain stereoscopic morphological data of the target building under the three-dimensional gradient; The execution module is used to obtain the temperature distribution of the target building, convert the three-dimensional morphological data into three-dimensional thermal infrared data according to the temperature distribution, and generate a three-dimensional thermal infrared model of the target building through the three-dimensional thermal infrared data.

7. A computer device, characterized in that: The computer device includes a memory and a processor, the memory is used to store a computer program, and the processor is used to call and run the computer program from the memory, so that the computer device executes the method for generating a three-dimensional building model according to any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores instructions or codes, and when the instructions or codes are executed on a computer, the computer implements the method for generating a three-dimensional building model as described in any one of claims 1 to 5.

Images