Typical residential building layout optimization design method and system in Tibet region

Abstract

The invention provides a Tibet typical residential building layout optimization design method and a Tibet typical residential building layout optimization design system, and relates to the technical field of image model optimization. Contour angle features of Tibet typical residential buildings are extracted from a point cloud image set, and filtering processing is performed on the point cloud image; obtaining a smooth descriptor of the typical residential building in the Tibet region during layout design; further determining a contour compensation factor of the typical residential building in the Tibet region during layout design according to the contour angle features and the smooth descriptor; performing feature reconstruction operation on all the point cloud images in the point cloud image set to obtain a feature fusion vector of the typical residential building in the Tibet region during layout design, and constructing a layout tracking model of the typical residential building in the Tibet region according to the feature fusion vector; according to the method, modeling compensation can be carried out on the contour error of the typical residential building in the Tibet region in the layout design process, so that the optimization efficiency of the layout design is improved.

Description

Technical Field

[0001] The present application relates to the technical field of image model optimization, and more specifically, to a typical residential building layout optimization design method and system thereof in Tibet. Background Art

[0002] Image model optimization technology is widely used in engineering design, architectural planning, geographic information systems and other fields. Its core lies in generating more accurate image models through the collection, processing, analysis and optimization modeling of image data, thereby improving the applicability and efficiency of the design. However, existing technologies often have problems such as data collection errors, noise interference and insufficient model building capabilities in complex terrain environments, such as plateau areas.

[0003] The layout design of typical residential buildings in Tibet needs to comprehensively consider the special environmental factors of the plateau. The digital twin technology can be combined to complete the optimization design of typical residential buildings in Tibet. However, the existing technology has failed to effectively solve the geometric distortion of point cloud data caused by complex terrain in the optimization design of the layout of typical residential buildings in Tibet, and lacks targeted contour error modeling and compensation mechanism. It is difficult to fully integrate terrain features, building contours and cultural features, which limits the overall optimization effect. Therefore, how to model and compensate for the contour errors of typical residential buildings in Tibet during the layout design process to improve the optimization efficiency of layout design has become a difficult problem faced by the industry. Summary of the invention

[0004] The present application provides a layout optimization design method and system for typical residential buildings in Tibet, which can model and compensate for contour errors of typical residential buildings in Tibet during the layout design process to improve the optimization efficiency of the layout design.

[0005] In the first aspect, the present application provides a typical residential building layout optimization design method in Tibet, the optimization design method comprising the following steps: Collect point cloud images of typical residential buildings in Tibet during layout design to obtain a point cloud image set; Extracting contour angle features of typical residential buildings in Tibet from the point cloud image set, filtering each point cloud image in the point cloud image set, and obtaining a smooth descriptor of the layout design of typical residential buildings in Tibet; Determine the contour compensation factor of the typical residential building in Tibet during layout design according to the contour angle feature and the smooth descriptor; Performing a feature reconstruction operation on all point cloud images in the point cloud image set, thereby obtaining a feature fusion vector of a typical residential building in Tibet during layout design, and constructing a layout tracking model of a typical residential building in Tibet based on the feature fusion vector; The layout tracking model is compensated and optimized based on the contour compensation factor.

[0006] In this embodiment, point cloud images of typical residential buildings in Tibet during layout design are collected through an infrared point cloud server.

[0007] In this embodiment, the point cloud image set is a collection of point cloud images.

[0008] In this embodiment, extracting the contour angle features of typical residential buildings in Tibet from the point cloud image set specifically includes: Performing convolution fusion on all point cloud images in the point cloud image set to obtain a contour map of typical residential buildings in Tibet; The contour map is subjected to histogram equalization, and then the contour angle features of typical residential buildings in Tibet are extracted from the contour map after the histogram equalization.

[0009] In this embodiment, filtering is performed on each point cloud image in the point cloud image set, and then a smooth descriptor of the layout design of typical residential buildings in Tibet is obtained, which specifically includes: Using Wiener filtering to perform pixel smoothing on each point cloud image in the point cloud image set, and obtaining a smoothed direction and a smoothed image corresponding to each point cloud image; Linear filtering is performed on all smoothed images according to the smoothing direction to obtain smoothing descriptors of typical residential buildings in Tibet during layout design.

[0010] In this embodiment, determining the contour compensation factor of a typical residential building in Tibet during layout design based on the contour angle feature and the smooth descriptor specifically includes: Get clear pixel outlines of typical residential buildings in Tibet; By using the contour angle feature and the clear pixel contour, heterogeneous matching is performed on the layout design process of typical residential buildings in Tibet to obtain a pixel matching error; Based on the pixel matching error and the smooth descriptor, a contour compensation factor of a typical residential building in Tibet during layout design is determined.

[0011] In this embodiment, the feature reconstruction operation is performed on all point cloud images in the point cloud image set, and the feature fusion vector of the layout design of typical residential buildings in Tibet is obtained, which specifically includes: Extracting the change amount of different image features of typical residential buildings in Tibet when the layout changes from the point cloud image set; Based on all the changes, the image features of all the point cloud images in the point cloud image set are enhanced and fused to obtain the feature fusion vector of the layout design of typical residential buildings in Tibet.

[0012] In this embodiment, constructing a layout tracking model of typical residential buildings in Tibet based on the feature fusion vector specifically includes: Acquire a cloud image set, and then determine tracking parameters based on the point cloud images in the cloud image set; Initialize the definition function of the image model; A layout tracking model of typical residential buildings in Tibet is determined based on the feature fusion vector, the tracking parameters and the definition function.

[0013] In this embodiment, performing compensation optimization on the layout tracking model based on the contour compensation factor refers to performing optimization registration on the layout tracking model based on a point cloud registration algorithm according to the contour compensation factor.

[0014] In a second aspect, the present application provides a typical residential building layout optimization design system in Tibet, which is used to implement a typical residential building layout optimization design method in Tibet. The optimization design system includes: An image acquisition module is used to collect point cloud images of typical residential buildings in Tibet during layout design to obtain a point cloud image set; An image processing module is used to extract the contour angle features of typical residential buildings in Tibet from the point cloud image set, and filter each point cloud image in the point cloud image set to obtain a smooth descriptor of the layout design of typical residential buildings in Tibet; The image processing module is also used to determine the contour compensation factor of the typical residential building in Tibet during layout design according to the contour angle feature and the smooth descriptor; The model building module is used to perform feature reconstruction operations on all point cloud images in the point cloud image set, thereby obtaining feature fusion vectors of typical residential buildings in Tibet during layout design, and constructing a layout tracking model of typical residential buildings in Tibet based on the feature fusion vectors.

[0015] A compensation optimization module is used to perform compensation optimization on the layout tracking model based on the contour compensation factor.

[0016] The technical solution provided by the embodiments disclosed in this application has the following beneficial effects: Firstly, point cloud images of typical residential buildings in Tibet during layout design are collected to obtain a point cloud image set; secondly, contour angle features of typical residential buildings in Tibet are extracted from the point cloud image set, and each point cloud image in the point cloud image set is filtered respectively to obtain a smooth descriptor of the typical residential buildings in Tibet during layout design; then, a contour compensation factor of the typical residential buildings in Tibet during layout design is determined according to the contour angle features and the smooth descriptor; then, a feature reconstruction operation is performed on all point cloud images in the point cloud image set to obtain a feature fusion vector of the typical residential buildings in Tibet during layout design, and a layout tracking model of the typical residential buildings in Tibet is constructed according to the feature fusion vector; finally, the layout tracking model is compensated and optimized based on the contour compensation factor.

[0017] It can be seen that in this application, the layout tracking model can be compensated and optimized through the contour compensation factor; first, by collecting point cloud images of typical residential buildings in Tibet, intuitive original data is provided for building layout design, and a data source is provided for accurately capturing building contours, terrain undulations and spatial distribution characteristics. Contour angle features are extracted from point cloud images, and smooth descriptors are generated in combination with filtering processing, which can effectively eliminate noise and interference in the collected data, thereby extracting key contour information, and significantly improving the clarity and availability of data in complex terrain and high-noise environments; secondly, the contour compensation factor is determined based on the contour angle features and the smooth descriptor, which provides core parameter support for subsequent error modeling. The contour compensation factor The determination of the sub-points can quantify and correct contour errors caused by complex terrain, distortion of point cloud data, etc., and by performing feature reconstruction operations on the point cloud image set and generating feature fusion vectors, the layout tracking model constructed based on the feature fusion vectors can reflect the comprehensive features in the actual scene; finally, the layout tracking model is compensated and optimized based on the contour compensation factor. By modeling and correcting the contour errors, it is beneficial to improve the accuracy of building layout design, ensure that the design results are still highly matched with actual needs in complex terrain environments, and achieve significant improvements in optimization efficiency; in summary, the present application scheme can model and compensate for the contour errors of typical residential buildings in Tibet during the layout design process, so as to improve the optimization efficiency of layout design. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] 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 the embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0019] Figure 1It is a flow chart of the optimization design method of typical residential building layout in Tibet provided by this application; Figure 2 is an exemplary flow chart for determining a smooth descriptor according to the present application; Figure 3 is an exemplary flow chart for determining a layout tracking model of typical residential buildings in Tibet provided by the present application; Figure 4 It is a module structure diagram of the typical residential building layout optimization design system in Tibet provided by this application. DETAILED DESCRIPTION

[0020] 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.

[0021] The embodiment of the present application provides a layout optimization design method and system for typical residential buildings in Tibet. The core of the method and system is as follows: first, point cloud images of typical residential buildings in Tibet during layout design are collected to obtain a point cloud image set; second, contour angle features of typical residential buildings in Tibet are extracted from the point cloud image set, and each point cloud image in the point cloud image set is filtered respectively to obtain a smooth descriptor of the typical residential buildings in Tibet during layout design; then, a contour compensation factor of the typical residential buildings in Tibet during layout design is determined according to the contour angle features and the smooth descriptor; then, a feature reconstruction operation is performed on all point cloud images in the point cloud image set to obtain a feature fusion vector of the typical residential buildings in Tibet during layout design, and a layout tracking model of the typical residential buildings in Tibet is constructed according to the feature fusion vector; finally, the layout tracking model is compensated and optimized based on the contour compensation factor.

[0022] Embodiment 1 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 As shown in FIG. 1 , this figure is an exemplary flow chart of a typical residential building layout optimization design method in Tibet according to this embodiment of the present application, and the optimization design method includes the following steps: In step S1, point cloud images of typical residential buildings in Tibet during layout design are collected to obtain a point cloud image set.

[0023] It should be noted that, in this embodiment, the point cloud images of typical residential buildings in Tibet during layout design can be collected through the infrared point cloud server to obtain a point cloud image set, wherein the point cloud image set is a collection of point cloud images; in specific implementation, the infrared images of typical residential buildings in Tibet during layout design can be collected through infrared image acquisition equipment, and the point cloud data of typical residential buildings in Tibet during layout design can be scanned using laser scanning equipment, and then the infrared point cloud server converts the infrared image and the point cloud data into a point cloud image, wherein the infrared image acquisition equipment can be an infrared camera, and the laser scanning equipment can be an infrared scanner. In other embodiments, other devices can also be used for acquisition, which is not limited here.

[0024] In step S2, the contour angle features of typical residential buildings in Tibet are extracted from the point cloud image set, and each point cloud image in the point cloud image set is filtered to obtain a smooth descriptor of the layout design of typical residential buildings in Tibet.

[0025] In this embodiment, the following method can be used to extract the contour angle features of typical residential buildings in Tibet from the point cloud image set, namely: Performing convolution fusion on all point cloud images in the point cloud image set to obtain a contour map of typical residential buildings in Tibet; The contour map is subjected to histogram equalization, and then the contour angle features of typical residential buildings in Tibet are extracted from the contour map after the histogram equalization.

[0026] In the specific implementation, first, the coordinates of each point cloud image in the point cloud image set are aligned to ensure that all point cloud images have the same scale, and the contour edges of typical residential buildings in Tibet are detected by an edge detection algorithm, and then a deep convolutional network is used to convolve all point cloud images based on the detected contour edges to obtain a contour map of typical residential buildings in Tibet, wherein the convolution kernel when convolving all point cloud images can be set to 3; then, the OpenCV library in the prior art is used to perform histogram equalization on the contour map, and the edge detection algorithm is used to extract the contour angle features of typical residential buildings in Tibet from the contour map after histogram equalization.

[0027] It should be noted that the contour angle feature in the present application refers to a set of geometric points in the external contour of a typical residential building in Tibet that reflect significant directional changes. The geometric points include corners, inflection points or points where the local curvature of the contour changes significantly. The contour angle feature includes local characteristics of the external contour of a typical residential building in Tibet, such as: direction, curvature and edge strength. The contour angle feature can reflect the key details of the building layout design process, which is conducive to identifying the key points of the external contour of a typical residential building in Tibet, thereby modeling and compensating for data errors and improving data accuracy.

[0028] Preferably, in this embodiment, reference Figure 2 As shown in FIG. 1 , this figure is an exemplary flow chart for determining a smooth descriptor in an embodiment of the present application. In this embodiment, each point cloud image in the point cloud image set is filtered respectively, and then a smooth descriptor of a typical residential building in Tibet during layout design is obtained, which can be specifically implemented by the following steps: First, in step S21, Wiener filtering is used to perform pixel smoothing on each point cloud image in the point cloud image set to obtain a smoothing direction and a smoothed image corresponding to each point cloud image; Then, in step S22, linear filtering is performed on all smoothed images according to the smoothing direction to obtain smoothing descriptors of typical residential buildings in Tibet during layout design.

[0029] It should be noted that in the present application, Wiener filtering is used to perform pixel smoothing on each point cloud image in the point cloud image set to obtain a smoothed image corresponding to each point cloud image, which is beneficial to the smoothing trend information of each area in the point cloud image and provides smoothed image data for subsequent filtering processing. In other embodiments, other existing technologies can also be used for smoothing processing, which is not limited here. In specific implementation, the smoothing direction represents the direction of pixel change. The gradient change direction of the pixels after filtering all point cloud images can be used as the smoothing direction. The smoothing direction can provide a direction basis for subsequent filtering processing.

[0030] In the specific implementation, a filter is constructed based on a Gaussian function to smooth all pixels of the smoothed image, the inverse of the smoothing direction is used as the smoothing direction of the filter, and a bilateral filter is generated by the filter, and then the bilateral filter is used as a smoothing descriptor for the layout design of typical residential buildings in Tibet. It should be noted that the Wiener filter performs pixel smoothing on each point cloud image in the point cloud image set to obtain a smoothed image corresponding to the point cloud image, and then a filter is constructed based on a Gaussian function to smooth all pixels of the smoothed image corresponding to the point cloud image, that is, each point cloud image is filtered twice.

[0031] It should be noted that the smoothness descriptor in the present application is an indicator for measuring the smoothness characteristics of image data pixels. The smoothness descriptor can represent the smoothness of the point cloud image. In the present embodiment, the smoothness descriptor is generated in pixels through Wiener filtering and Gaussian filtering, which is beneficial to reduce the noise of the point cloud image and describe the edge features and directional characteristics of the building contour, facilitate feature extraction and model optimization, and provide an accurate description of the contour error of typical residential buildings in Tibet during layout design.

[0032] In step S3, the contour compensation factor of the typical residential building in Tibet during layout design is determined according to the contour angle feature and the smoothness descriptor.

[0033] In this embodiment, the contour compensation factor of typical residential buildings in Tibet during layout design can be determined based on the contour angle feature and the smooth descriptor in the following manner, namely: Get clear pixel outlines of typical residential buildings in Tibet; By using the contour angle feature and the clear pixel contour, heterogeneous matching is performed on the layout design process of typical residential buildings in Tibet to obtain a pixel matching error; Based on the pixel matching error and the smooth descriptor, a contour compensation factor of a typical residential building in Tibet during layout design is determined.

[0034] In specific implementation, the digital twin technology can be used to simulate and generate clear pixel contours of typical residential buildings in Tibet, provide basic analysis data of the actual building appearance, and facilitate matching contour errors; in addition, the pixel matching error is an indicator to measure the degree of contour error of the point cloud image, and the heterogeneous matching is to match image data acquired in different ways, which is beneficial to reduce errors during image processing. The pixel matching error can be determined in the following way, namely: first, the descriptor algorithm is used to match the feature descriptors based on Euclidean distance for the point coordinates in the contour angle feature; then, the Euclidean distance between each point coordinate in the contour angle feature and the matching point coordinate in the clear pixel contour is used as the matching error, and then the matching error of all point coordinates is obtained; finally, the average of all matching errors is used as the pixel matching error.

[0035] In addition, in a specific implementation, the contour compensation factor of a typical residential building in Tibet during layout design can be determined by the absolute value of the smooth descriptor and the pixel matching error. The contour compensation factor is a quantized parameter used to correct the contour error in the layout design of a typical residential building in Tibet. The contour compensation factor compensates and optimizes the geometric deviations that occur during the layout design process by combining the smooth descriptor and the pixel matching error, which is conducive to constructing an accurate building contour model. In actual implementation, the weighted sum of the absolute value of the smooth descriptor and the pixel matching error can be used as the contour compensation factor of a typical residential building in Tibet during layout design.

[0036] It should be noted that, in this embodiment, the weight value of the weighted sum of the absolute value of the smooth descriptor and the pixel matching error can be determined in the following manner, namely: first, the weights of the absolute value of the smooth descriptor and the pixel matching error are both set to 0.5, and then, a weight training model is constructed based on a machine learning algorithm; finally, all pixels of the point cloud image, the absolute value of the smooth descriptor and the pixel matching error are used as inputs of the weight training model, and then the weight training model outputs the weight value of the weighted sum of the absolute value of the smooth descriptor and the pixel matching error. Among them, a decision tree algorithm can be used to construct a weight training model based on a machine learning algorithm, which is beneficial to improve the reliability of the output weight value. Other machine learning algorithms can also be used in other embodiments, which are not limited here.

[0037] In step S4, a feature reconstruction operation is performed on all point cloud images in the point cloud image set to obtain a feature fusion vector of a typical residential building in Tibet during layout design, and a layout tracking model of a typical residential building in Tibet is constructed based on the feature fusion vector.

[0038] In this embodiment, the feature reconstruction operation is performed on all point cloud images in the point cloud image set, and the feature fusion vector of the typical residential building in Tibet during the layout design is obtained in the following manner, namely: Extracting the change amount of different image features of typical residential buildings in Tibet when the layout changes from the point cloud image set; Based on all the changes, the image features of all the point cloud images in the point cloud image set are enhanced and fused to obtain the feature fusion vector of the layout design of typical residential buildings in Tibet.

[0039] It should be noted that the different image features of typical residential buildings in Tibet when the layout changes include color, texture, shape and other image features. The mean of the difference in image features between all point cloud images when the layout of typical residential buildings in Tibet changes can be used as the change of the corresponding image feature, and then the change of all image features can be obtained.

[0040] In specific implementation, the changes in different image features can be normalized, and the main plant analysis method can be used to screen out important image features. The changes in the screened image features are then used as the allocation index of feature weights. Based on the feature weights, all the screened image features are weightedly fused to obtain the feature fusion vector of the layout design of typical residential buildings in Tibet, where the direction of the feature fusion vector is random.

[0041] It should be noted that the feature fusion vector in the present application is a feature quantity that describes the changes in the image features of typical residential buildings in Tibet during layout design. The feature fusion vector comprehensively considers the changes in multiple image features such as color, texture, and shape, so as to reflect the overall image features of typical residential buildings in Tibet under different layout changes, thereby facilitating capturing changes in the appearance of buildings during layout design.

[0042] Preferably, in this embodiment, reference Figure 3 As shown, this figure is an exemplary flow chart for determining the layout tracking model of typical residential buildings in Tibet in an embodiment of the present application. In this embodiment, the layout tracking model of typical residential buildings in Tibet is constructed based on the feature fusion vector.

[0043] First, in step S41, a cloud image set is obtained, and then tracking parameters are determined according to the point cloud images in the cloud image set; Then, in step S42, the definition function of the image model is initialized; Finally, in step S43, a layout tracking model of typical residential buildings in Tibet is determined based on the feature fusion vector, the tracking parameters and the definition function.

[0044] In the specific implementation, first, the point cloud image in the cloud image set can be regarded as a continuous changing frame image, and the outline of the typical residential building in Tibet can be identified by the edge detection algorithm, and then the tracking parameters in the identification process are extracted; then, a definition function is initialized based on the weighted feature fusion algorithm, and the definition function can be expressed as ; Finally, the pixels of the point cloud image are read, and the layout tracking model of the typical residential building in Tibet is determined by reading the pixels, the feature fusion vector, the tracking parameter and the definition function, wherein the layout tracking model of the typical residential building in Tibet can be defined by the following expression:

[0045] in, An expression representing the layout tracking model, Represents the weight parameter, which can be determined by building a weight training model based on a machine learning algorithm. represents the feature fusion vector, Represents the tracking parameters, The pixels of the point cloud image can be used to obtain the layout tracking model of the typical residential buildings in Tibet through the above expression using the visualization library in the prior art; it should be noted that the layout tracking model in the present application is an image model, which can be determined based on the model expression using the visualization library in the prior art, and the layout tracking model can represent the layout changes of the typical residential buildings in Tibet at different time points for real-time feedback, and can reflect the image characteristics of the typical residential buildings in Tibet. In addition, the layout tracking model can also accurately describe the layout changes of the buildings.

[0046] In step S5, compensation optimization is performed on the layout tracking model based on the contour compensation factor.

[0047] It should be noted that, in the present embodiment, compensating and optimizing the layout tracking model based on the contour compensation factor refers to optimizing and registering the layout tracking model based on the point cloud registration algorithm according to the contour compensation factor, wherein the contour compensation factor adjusts the parameters of the model according to the contour features of the building in different layout states, so that the model is more in line with the changes in the actual layout. In the present embodiment, the point cloud data set obtained by the point cloud registration algorithm is aligned to generate an accurate image model, which is conducive to accurately reflecting the spatial structure of typical residential buildings in Tibet.

[0048] It can be seen that in this application, the layout tracking model can be compensated and optimized by the contour compensation factor; first, by collecting point cloud images of typical residential buildings in Tibet, intuitive raw data is provided for building layout design, and a data source is provided for accurately capturing building contours, terrain undulations and spatial distribution characteristics. Contour angle features are extracted from point cloud images, and smooth descriptors are generated in combination with filtering processing, which can effectively eliminate noise and interference in the collected data, thereby extracting key contour information, and significantly improving the clarity and availability of data in complex terrain and high-noise environments; secondly, the contour compensation factor is determined based on the contour angle features and the smooth descriptor, which provides core parameter support for subsequent error modeling. The determination of the contour compensation factor can quantify and correct contour errors caused by complex terrain, point cloud data distortion and other reasons. By performing feature reconstruction operations on the point cloud image set and generating feature fusion vectors, the layout tracking model constructed based on the feature fusion vector can reflect the comprehensive features in the actual scene; finally, the layout tracking model is compensated and optimized based on the contour compensation factor. By modeling and correcting the contour error, it is beneficial to improve the accuracy of the building layout design, ensure that the design results can still be highly matched with actual needs in a complex terrain environment, and achieve a significant improvement in optimization efficiency.

[0049] In summary, the present application scheme can model and compensate for the contour errors of typical residential buildings in Tibet during the layout design process, so as to improve the optimization efficiency of the layout design.

[0050] Embodiment 2 This application provides a typical residential building layout optimization design system in Tibet. Figure 4 As shown in FIG. 1 , this figure is a module structure diagram of a typical residential building layout optimization design system in Tibet provided by this application, and the optimization design system includes: The image acquisition module 100 is used to acquire point cloud images of typical residential buildings in Tibet during layout design to obtain a point cloud image set; The image processing module 200 is used to extract the contour angle features of typical residential buildings in Tibet from the point cloud image set, and filter each point cloud image in the point cloud image set to obtain a smooth descriptor of the layout design of typical residential buildings in Tibet; The image processing module 200 in the present application is also used to determine the contour compensation factor of the typical residential building in Tibet during layout design according to the contour angle feature and the smooth descriptor; The model building module 300 is used to perform feature reconstruction operations on all point cloud images in the point cloud image set, thereby obtaining feature fusion vectors of typical residential buildings in Tibet during layout design, and constructing a layout tracking model of typical residential buildings in Tibet based on the feature fusion vectors.

[0051] The compensation optimization module 400 is used to perform compensation optimization on the layout tracking model based on the contour compensation factor.

[0052] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems) and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0053] A person skilled in the art may understand that all or part of the steps in the various methods of the above embodiments may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, the storage medium including a read-only memory (ROM), a random access memory (RAM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), a one-time programmable read-only memory (OTPROM), an electronically-erasable programmable read-only memory (EEPROM), a compact disc (CD-ROM) or other optical disc storage, magnetic disk storage, magnetic tape storage, or any other computer-readable medium that can be used to carry or store data.

[0054] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

Claims

1. The optimization design method of typical residential building layout in Tibet is characterized by: The optimization design method comprises the following steps: Collect point cloud images of typical residential buildings in Tibet during layout design to obtain a point cloud image set; Extracting contour angle features of typical residential buildings in Tibet from the point cloud image set, filtering each point cloud image in the point cloud image set, and obtaining a smooth descriptor of the layout design of typical residential buildings in Tibet; Determine the contour compensation factor of the typical residential building in Tibet during layout design according to the contour angle feature and the smooth descriptor; Performing a feature reconstruction operation on all point cloud images in the point cloud image set, thereby obtaining a feature fusion vector of a typical residential building in Tibet during layout design, and constructing a layout tracking model of a typical residential building in Tibet based on the feature fusion vector; The layout tracking model is compensated and optimized based on the contour compensation factor.

2. The method for optimizing the layout of typical residential buildings in Tibet as claimed in claim 1, characterized in that: Point cloud images of typical residential buildings in Tibet during layout design are collected through the infrared point cloud server.

3. The method for optimizing the layout of typical residential buildings in Tibet as claimed in claim 1, characterized in that: The point cloud image set is a set of point cloud images.

4. The method for optimizing the layout of typical residential buildings in Tibet as claimed in claim 1, characterized in that: Extracting the contour angle features of typical residential buildings in Tibet from the point cloud image set specifically includes: Performing convolution fusion on all point cloud images in the point cloud image set to obtain a contour map of typical residential buildings in Tibet; The contour map is subjected to histogram equalization, and then the contour angle features of typical residential buildings in Tibet are extracted from the contour map after the histogram equalization.

5. The method for optimizing the layout of typical residential buildings in Tibet as claimed in claim 1, characterized in that: Each point cloud image in the point cloud image set is filtered respectively, and then the smooth descriptor of the layout design of typical residential buildings in Tibet is obtained, which specifically includes: Using Wiener filtering to perform pixel smoothing on each point cloud image in the point cloud image set, and obtaining a smoothed direction and a smoothed image corresponding to each point cloud image; Linear filtering is performed on all smoothed images according to the smoothing direction to obtain smoothing descriptors of typical residential buildings in Tibet during layout design.

6. The method for optimizing the layout of typical residential buildings in Tibet as claimed in claim 1, characterized in that: Determining the contour compensation factor of typical residential buildings in Tibet during layout design based on the contour angle feature and the smooth descriptor specifically includes: Get clear pixel outlines of typical residential buildings in Tibet; By using the contour angle feature and the clear pixel contour, heterogeneous matching is performed on the layout design process of typical residential buildings in Tibet to obtain a pixel matching error; Based on the pixel matching error and the smooth descriptor, a contour compensation factor of a typical residential building in Tibet during layout design is determined.

7. The method for optimizing the layout of typical residential buildings in Tibet as claimed in claim 1, characterized in that: Performing feature reconstruction operations on all point cloud images in the point cloud image set, and then obtaining feature fusion vectors of typical residential buildings in Tibet during layout design specifically includes: Extracting the change amount of different image features of typical residential buildings in Tibet when the layout changes from the point cloud image set; Based on all the changes, the image features of all the point cloud images in the point cloud image set are enhanced and fused to obtain the feature fusion vector of the layout design of typical residential buildings in Tibet.

8. The method for optimizing the layout of typical residential buildings in Tibet as claimed in claim 1, characterized in that: The layout tracking model of typical residential buildings in Tibet is constructed based on the feature fusion vector, which specifically includes: Acquire a cloud image set, and then determine tracking parameters based on the point cloud images in the cloud image set; Initialize the definition function of the image model; A layout tracking model of typical residential buildings in Tibet is determined based on the feature fusion vector, the tracking parameters and the definition function.

9. The method for optimizing the layout of typical residential buildings in Tibet as claimed in claim 1, characterized in that: Compensating and optimizing the layout tracking model based on the contour compensation factor refers to optimizing and registering the layout tracking model based on a point cloud registration algorithm according to the contour compensation factor.

10. A typical residential building layout optimization design system in Tibet, used to implement the typical residential building layout optimization design method in Tibet as claimed in any one of claims 1 to 9, characterized in that: The optimization design system comprises: An image acquisition module is used to collect point cloud images of typical residential buildings in Tibet during layout design to obtain a point cloud image set; An image processing module is used to extract the contour angle features of typical residential buildings in Tibet from the point cloud image set, and filter each point cloud image in the point cloud image set to obtain a smooth descriptor of the layout design of typical residential buildings in Tibet; The image processing module is also used to determine the contour compensation factor of the typical residential building in Tibet during layout design according to the contour angle feature and the smooth descriptor; The model building module is used to perform feature reconstruction operations on all point cloud images in the point cloud image set, thereby obtaining feature fusion vectors of typical residential buildings in Tibet during layout design, and constructing a layout tracking model of typical residential buildings in Tibet based on the feature fusion vectors. A compensation optimization module is used to perform compensation optimization on the layout tracking model based on the contour compensation factor.

Images