An automatic measurement method of urban form based on the fourth-order spatial entropy matrix model

Abstract

The present invention discloses an automatic urban morphology measurement method based on a spatial entropy fourth-order matrix model, which relates to the field of urban design technology. The method performs fourth-order matrix modeling based on the relative morphological information of the target building and surrounding buildings, and then classifies the buildings according to the clustering algorithm of the fourth-order matrix. The city is partitioned according to the spatial entropy value measurement, thereby identifying problematic buildings and urban areas. The present invention can meet the work needs of large-scale urban renewal and automatically generate a morphological transformation plan specific to each building. It has the characteristics of large scale and high precision, thereby improving the work efficiency of urban renewal.

Description

Technical Field

[0001] The present invention relates to the technical field of urban design, and in particular to an automatic urban form measurement method based on a spatial entropy fourth-order matrix model. Background Art

[0002] Traditional urban morphology measurement research is mainly carried out through surveys, images and model studies. Although the measurement methods of urban morphology have achieved a transformation from measuring single morphological elements to integrating the measurement of multiple urban morphological elements, most of them start from the study of the relationship organization of streets and blocks in the city and the properties of the buildings themselves, and lack the perspective of the impact of the relationship between buildings on urban morphology. However, according to statistical research, the impact of the organizational relationship between buildings on urban morphology is three times stronger than the impact of the organizational relationship between blocks and streets on urban morphology. Summary of the Invention

[0003] In order to solve the deficiencies mentioned in the above background technology, the purpose of the present invention is to provide an automatic urban morphology measurement method based on a spatial entropy fourth-order matrix model, which can measure urban morphology according to the relationships between buildings.

[0004] The purpose of the present invention can be achieved by the following technical solution: an automatic urban form measurement method based on a spatial entropy fourth-order matrix model, the method comprising the following steps:

[0005] The target buildings in the urban area are numbered, and the four morphological indicators of the plane distance, height difference, orientation difference, and volume difference between each target building and the surrounding buildings are measured and calculated using a total station. The coordinate information of the plane center point of each target building is obtained using a GPS measuring instrument;

[0006] A fourth-order matrix is constructed using four morphological indicators: plane distance, height difference, orientation difference, and volume difference. The building is then modeled using the fourth-order matrix to generate a fourth-order matrix model. A supervised clustering algorithm is used to train the fourth-order matrix model of the building, resulting in a fourth-order matrix intelligent clustering model.

[0007] A fourth-order matrix intelligent clustering model is used to classify building morphology and identify landmark buildings, shanty buildings, and buildings with insufficient daylight. Combined with the coordinates of the center point of the target building plane, the diversity of the fourth-order matrix of buildings in a certain zone in the area is used as the spatial entropy measurement indicator of the zone. Urban areas are divided based on the lowest spatial entropy value of the zone, and morphologically chaotic areas, morphologically monotonous areas, and change fault zones in the city are identified.

[0008] The object building classification and urban area zoning results are displayed in the form of point clouds of different colors. Multiple rounds of cyclic optimization are performed to display the optimized fourth-order matrix information of the object building, and the height, volume, orientation and coordinate information of the object building after modifying the fourth-order matrix information are obtained.

[0009] Preferably, the process of numbering the target buildings in the urban area, measuring and calculating four morphological index data of the plane distance, height difference, orientation difference, and volume difference between each target building and the surrounding buildings by a total station, and obtaining the coordinate information of the plane center point of the target building by a GPS measuring instrument includes the following steps:

[0010] Each building in the urban area is assigned a 14-digit code based on the following principles: the first six digits are the postal code of the administrative district where the building is located, the middle four digits are the block number, and the last four digits are the building number within the block;

[0011] Using a total station, the relative data of the plane distance, height difference, orientation difference and volume difference between the target building and the surrounding buildings are obtained one by one according to the coding sequence;

[0012] Using a GPS measuring instrument, the coordinates of the center points of the building plane are obtained one by one according to the coding sequence, and the center point coordinates are exported as ASCII coding format.

[0013] Preferably, the plane distance between the target building and the surrounding buildings is measured using the precision measurement mode of the total station with a minimum display unit of 1mm; the vertical angle difference between the target building and the surrounding buildings is measured using the horizontal angle measurement mode of the total station with a minimum angle measurement accuracy of 0.5°, and the height difference is obtained by combining the trigonometric height method; the orientation difference between the target building and the surrounding buildings is measured using the azimuth measurement mode of the total station with a minimum angle measurement accuracy of 0.5°; the volume difference between the target building and the surrounding buildings is obtained in the earthwork measurement mode of the total station, and the four types of relative data are exported in DAT format.

[0014] Preferably, the process of deriving the fourth-order matrix intelligent clustering model includes the following steps:

[0015] Using the obtained four morphological index data of plane distance, height difference, orientation difference and volume difference, a fourth-order matrix model is generated for each target building in the urban area. The four morphological index data are then linked to the fourth-order matrix model for display. The first row of the fourth-order matrix describes the plane distance between the target building and the surrounding buildings, the second row of the fourth-order matrix describes the height difference between the target building and the surrounding buildings, the third row of the fourth-order matrix describes the orientation difference between the target building and the surrounding buildings, and the fourth row of the fourth-order matrix describes the volume difference between the target building and the surrounding buildings.

[0016] The four morphological index data of the fourth-order matrix model are used as machine learning labels. The target fourth-order matrix is divided into a training set, a validation set, and a test set in a ratio of 98:1:1. A holographic sand table is used to demonstrate the fourth-order matrix of all buildings and building links in the urban area. Similar buildings of each building are found one by one on the holographic sand table, and the connection information is recorded and stored.

[0017] A supervised clustering learning algorithm is used to perform fourth-order matrix clustering machine learning training through a graphics memory deep learning system. A machine learning model with strong generalization performance is selected as the fourth-order matrix automatic clustering model through cross-validation and generalization test.

[0018] Preferably, the number of columns of the fourth-order matrix is determined by the number of surrounding buildings of the target building.

[0019] Preferably, the process of classifying the morphology of the target building using the fourth-order matrix intelligent clustering model includes the following steps:

[0020] The fourth-order matrix information of the target buildings in the urban area is classified using a fourth-order matrix automatic clustering model consisting of the distance difference, height difference, orientation difference and volume difference between the target building and the surrounding buildings;

[0021] The fourth-order matrix information classified by the fourth-order matrix automatic clustering model is tested, among which the fourth-order matrix plane distance difference information value, height difference information value and volume difference information value of landmark buildings should be more than 20% higher than the average of other buildings; the fourth-order matrix plane distance difference information value and height difference information value of shanty buildings should be more than 20% lower than the average of other buildings; the fourth-order matrix plane distance difference information value and height difference information value of buildings with insufficient lighting should be more than 20% lower than the average of other buildings; when errors occur in the identification test results, the process of fourth-order matrix clustering machine learning training is repeated through the video memory deep learning system until the building information results classified by the automatic clustering model meet the test and identification requirements.

[0022] Preferably, the process of partitioning the urban area based on the coordinates of the center point of the target building plane and the diversity of the fourth-order building matrix of a certain partition in the area as the spatial entropy measurement index of the partition and taking the lowest spatial entropy value of the partition as the standard comprises the following steps:

[0023] Cluster analysis is performed based on the coordinates of the center points of the building planes. Buildings with similar plane coordinates are divided into a zone, and the diversity of the fourth-order matrix changes in the zone is used as the measurement standard of spatial entropy. The measurement standard threshold of spatial entropy is set at 20%. If it is higher than the measurement standard threshold of spatial entropy, the morphological zone is reduced to 0.8 times of the original zone based on the classification results of the fourth-order matrix information of the building. Object buildings of a smaller number of classification types in the zone are eliminated, and then the zone is re-zoned. The morphological zone is compared with the spatial entropy measurement standard threshold again until the spatial entropy measurement standard range requirements are met. The regional morphological zone is optimized and the final regional morphological zone result is obtained. The calculation formula of spatial entropy is:

[0024] H(x)=-∑p(xi)logp(xi)

[0025] Where H(x) is the spatial entropy of partition x, and p(xi) is the ratio of the number of building type i in partition x to the total number of buildings in the partition.

[0026] Identify morphologically chaotic areas, morphologically monotonous areas and change fault zones in urban areas. The basis for identifying morphologically chaotic areas is that the spatial entropy measurement value in the area is more than 3 times higher than that in other areas; the basis for identifying morphologically monotonous areas is that the spatial entropy measurement value in the area is less than 30% of that in other areas; and the change fault zone is that the fluctuation range of the spatial entropy measurement value of the adjacent surrounding areas is greater than 50%.

[0027] Preferably, the process of displaying the object building classification and urban area zoning results in the form of point clouds of different colors, performing multiple rounds of cyclic optimization, displaying the optimized fourth-order matrix information of the object building, and obtaining the height, volume, orientation and coordinate information of the object building after modifying the fourth-order matrix information includes the following steps:

[0028] The results of building classification and urban area zoning are displayed in the form of point clouds of different colors. The fourth-order matrix information of some buildings is modified based on six judgment rules: retaining landmark buildings, reducing shanty buildings, buildings with insufficient daylight, areas with chaotic morphology, areas with monotonous morphology, and changing fault zones.

[0029] Optimize the fourth-order matrix information of the target building in a loop until the calculated values of the six types of judgment rules no longer change. Then output the modified fourth-order matrix information and classification and partitioning results, and display them in the holographic sandbox.

[0030] Based on the modified fourth-order matrix information, the updated building height, volume, orientation and coordinate information are derived.

[0031] Preferably, a device comprises:

[0032] one or more processors;

[0033] a memory for storing one or more programs;

[0034] When one or more of the programs are executed by one or more of the processors, the one or more processors implement any of the above-described methods for automatic measurement of urban form based on a fourth-order spatial entropy matrix model.

[0035] Preferably, a storage medium comprising computer executable instructions, wherein the computer executable instructions, when executed by a computer processor, are used to execute any of the above-described methods for automatic measurement of urban form based on a fourth-order spatial entropy matrix model.

[0036] Beneficial effects of the present invention:

[0037] The fourth-order matrix automatic clustering model obtained by machine learning in this invention can take into account the morphological relationships between buildings more than existing classification methods. The application of this model in the field of building classification allows multi-line parallel calculation of building classification, which doubles the efficiency of building classification and improves the classification accuracy to 85%, which will help the subsequent optimization and updating of building and urban morphology.

[0038] The present invention measures urban morphology through spatial entropy, which is different from previous subjective judgments and can more objectively define urban morphological zones, which is conducive to the subsequent optimization of urban morphology.

[0039] During the fourth-order matrix cyclic optimization stage, the present invention links the fourth-order matrix information of the building to the urban model for display, which can correspond the urban form and the updated effect in real time, making the modification process more intuitive, shortening the decision-making time by 61% and facilitating diversified participation in the optimization process.

[0040] The present invention can further output an urban renewal building renovation worksheet, refine the conceptual update guidelines of traditional pictures, videos, etc., and obtain specific update data of building height, volume, orientation, and coordinate information through the derivation of the fourth-order matrix. At the same time, combined with the morphological volume diagram, subsequent update operations are more intuitive and the update process is more efficient and controllable. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, those skilled in the art can derive other drawings based on these drawings without inventive effort.

[0042] Figure 1 is a flow chart of the method of the present invention;

[0043] Figure 2 It is a schematic diagram of the fourth-order matrix model of the building;

[0044] Figure 3 This is a diagram of data collection for the fourth-order matrix before building update in Example A. DETAILED DESCRIPTION

[0045] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts shall fall within the scope of protection of the present invention.

[0046] like Figure 1 As shown, an automatic urban form measurement method based on a fourth-order spatial entropy matrix model includes the following steps:

[0047] (1) The buildings in the central urban area of Fuzhou were uniformly numbered. An intelligent total station with a 33X magnification and a 2.5-resolution telescope was used to measure and calculate the four morphological indicators of the plane distance, height difference, orientation difference, and volume difference between each building in the central urban area of Fuzhou and its surrounding buildings. A GPS measuring instrument with an accuracy of 0.1m was used to obtain the coordinate information of the plane center point of the buildings in the central urban area.

[0048] (1.1) Each building in the central urban area of Fuzhou is assigned a fourteen-digit code. The coding is based on the following: the first six digits are the postal code of the administrative district in Fuzhou where the building is located (Gulou District: 350001; Taijiang District: 350004; Cangshan District: 350007; Mawei District: 350015; Jin'an District: 350011), the middle four digits are the block number, and the last four digits are the building number within the block.

[0049] (1.2) Using an intelligent total station equipped with a 33X magnification and 2.5 resolution telescope, the relative morphological data of the plane distance, height difference, orientation difference and volume difference between each building in the central urban area of Fuzhou and its surrounding buildings are obtained one by one in the sequence described in (1.1). Among them, the plane distance between the building and the surrounding buildings is measured using the precision measurement mode of the total station with a minimum display unit of 1mm; the vertical angle difference between the building and the surrounding buildings is measured using the horizontal angle measurement mode of the total station with a minimum angular measurement accuracy of 0.5°, and the height difference is obtained by combining the trigonometric height method; the orientation difference between the building and the surrounding buildings is measured using the azimuth measurement mode of the total station with a minimum angular measurement accuracy of 0.5°; the volume difference between the building and the surrounding buildings is obtained using the earthwork measurement mode of the total station, and the four relative morphological data are exported in DAT format and entered into the computer.

[0050] (1.3) Using a GPS measuring instrument with an accuracy of 0.1 m, obtain the plane center point coordinates of each building in the central urban area of Fuzhou one by one in the sequence described in (1.1), and export the center point coordinates in ASCII format and enter them into the computer.

[0051] (2) Using the four indicators of plane distance, height difference, orientation difference, and volume difference to construct a fourth-order matrix, and then modeling is performed to obtain a fourth-order matrix model, such as Figure 2 As shown. A supervised clustering algorithm is used to perform machine learning training on the fourth-order matrix model of the building to obtain a fourth-order matrix intelligent clustering model, which specifically includes:

[0052] (2.1) Using the obtained four morphological index data of plane distance, height difference, orientation difference and volume difference, a fourth-order matrix model is constructed for each building in the area, and the matrix is linked to the model for display. Taking building A as an example, the first row of the fourth-order matrix describes the plane distance between the target building and the surrounding buildings, the second row describes the height difference between the target building and the surrounding buildings, the third row describes the orientation difference between the target building and the surrounding buildings, and the fourth row describes the volume difference between the target building and the surrounding buildings. The specific number of columns is 7, as shown in the following example: Figure 3 shown.

[0053] (2.2) The four basic elements of plane distance, height difference, orientation difference, and volume difference are used as machine learning labels. The target fourth-order matrix is divided into training set, validation set, and test set in a ratio of 98:1:1. A holographic sand table is used to demonstrate the buildings in the area and the fourth-order matrix of their links. The operator wears data gloves to select similar buildings and their corresponding matrices for the demonstrated buildings, and the operator's selection tendency is obtained with the assistance of an eye tracker.

[0054] (2.3) Use the deep learning system to perform fourth-order matrix clustering machine learning training, and select a machine learning model with strong generalization performance as the fourth-order matrix automatic clustering model through the K-fold cross-validation method.

[0055] (3) A fourth-order matrix intelligent clustering model was used to classify the architectural forms in the central urban area of Fuzhou City, and to identify the key landmark buildings, shanty buildings, and buildings with insufficient daylighting in Fuzhou City. Combined with the coordinates of the center points of the building planes in the central urban area, the central urban area of Fuzhou City was divided into regions according to the results of regional spatial entropy measurement, and the chaotic morphological areas, monotonous morphological areas, and change fault zones in the central urban area were identified.

[0056] (3.1) Cluster the fourth-order matrix information of the buildings in the central urban area of Fuzhou obtained in step (1) using a fourth-order matrix automatic clustering model consisting of the distance difference, height difference, orientation difference and volume difference between adjacent buildings.

[0057] (3.2) The fourth-order matrix information of buildings in the central urban area of Fuzhou classified by the fourth-order matrix automatic clustering model is tested. Among them, the fourth-order matrix plane distance difference information value, height difference information value and volume difference information value of landmark buildings should be higher than the average of other buildings by more than 20%; the fourth-order matrix plane distance difference information value and height difference information value of shanty buildings should be lower than the average of other buildings by more than 20%; the fourth-order matrix plane distance difference information value and height difference information value of buildings with insufficient lighting should be lower than the average of other buildings by more than 20%; when the recognition test results produce errors, the process of fourth-order matrix clustering machine learning training is repeated through the video memory deep learning system until the building information results classified by the automatic clustering model meet the test and recognition requirements.

[0058] (3.3) Cluster analysis was performed based on the center point coordinates of the building planes in the central urban area of Fuzhou City. Buildings with similar plane coordinates were divided into a zone, and the diversity of the fourth-order matrix within the zone was used as the measurement standard for spatial entropy. The measurement standard threshold of spatial entropy was set at 20%. If it was higher than the measurement standard threshold of spatial entropy, the morphological zone was reduced to 0.8 times the original zone based on the classification results of the building fourth-order matrix information. Object buildings of the classification type with a small number in the zone were eliminated, and then the zone was re-zoned. The morphological zone was compared with the spatial entropy measurement standard threshold again until the spatial entropy measurement standard range requirements were met. The regional morphological zone was optimized and the final regional morphological zone result was obtained. The calculation formula for spatial entropy is:

[0059] H(x)=-∑p(xi)logp(xi)

[0060] Where H(x) is the spatial entropy of partition x, and p(xi) is the ratio of the number of building type i in partition x to the total number of buildings in the partition.

[0061] (3.4) Identify the morphologically chaotic areas, morphologically monotonous areas, and variable fault zones within the central urban area. The morphologically chaotic areas are identified when the spatial entropy measurement value within the area is more than three times higher than that of other areas; the morphologically monotonous areas are identified when the spatial entropy measurement value within the area is less than 30% lower than that of other areas; and the variable fault zones are identified when the spatial entropy measurement value of the adjacent surrounding areas fluctuates by more than 50%.

[0062] (4) Display the building classification and regional zoning results in the form of point clouds of different colors on the geographic information platform, modify the fourth-order matrix information of some buildings, and perform multiple rounds of cyclic optimization; display the optimized fourth-order matrix information of urban form, and generate and print the "Urban Renewal Building Reconstruction Worksheet" according to the building number.

[0063] (4.1) The resulting building classification and regional zoning results are displayed in the form of colored point clouds on the geographic information platform. A holographic sand table is used to display the city model linked to the fourth-order matrix information of the urban morphology. Users wearing data gloves connect to the 3D model processing software and interact with the holographic sand table. The fourth-order matrix information of some buildings is modified according to six judgment rules: retaining landmark buildings, reducing shanty buildings, buildings with insufficient daylight, areas with chaotic morphology, areas with monotonous morphology, and changing fault zones.

[0064] (4.2) Optimize the fourth-order matrix information of the building in a loop until the calculated values of the six categories of judgment rules no longer change significantly, and then output the modified fourth-order matrix information and classification and partitioning results.

[0065] (4.3) Use the geographic information platform to display the final results of building classification and regional zoning. Based on the modified matrix information, derive the updated building height, volume, orientation, and coordinate information. Based on the building code, connect a printer to print the "Urban Renewal Building Renovation Worksheet." Using Building A as an example, the specific contents are shown in Table 1.

[0066] Table 1 Urban Renewal Building Renovation Worksheet

[0067]

[0068]

[0069] Based on the same inventive concept, the present invention also provides a computer device, which includes: one or more processors, and a memory for storing one or more computer programs; the program includes program instructions, and the processor is used to execute the program instructions stored in the memory. The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. It is the computing core and control core of the terminal, which is used to implement one or more instructions, specifically for loading and executing one or more instructions in a computer storage medium to implement the above method.

[0070] It should be further explained that, based on the same inventive concept, the present invention also provides a computer storage medium having a computer program stored thereon, which executes the above method when executed by a processor. The storage medium can be any combination of one or more computer-readable media. The computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electrical, magnetic, infrared, or semiconductor system, device or component, or any combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present invention, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, device or component.

[0071] Throughout this specification, references to terms such as "one embodiment," "example," or "specific example" indicate that a specific feature, structure, material, or characteristic described in conjunction with that embodiment or example is included in at least one embodiment or example of the present disclosure. In this specification, schematic representations of these terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

[0072] The above shows and describes the basic principles, main features and advantages of the present disclosure. Those skilled in the art should understand that the present disclosure is not limited to the above embodiments. The above embodiments and descriptions are merely illustrative of the principles of the present disclosure. Various changes and improvements may be made to the present disclosure without departing from the spirit and scope of the present disclosure, and such changes and improvements shall fall within the scope of the present disclosure.

Claims

1. An automatic urban form measurement method based on a fourth-order spatial entropy matrix model, characterized by: The method comprises the following steps: The target buildings in the urban area are numbered, and the four morphological indicators of the plane distance, height difference, orientation difference, and volume difference between each target building and the surrounding buildings are measured and calculated using a total station. The coordinate information of the plane center point of each target building is obtained using a GPS measuring instrument; A fourth-order matrix is constructed using four morphological indicators: plane distance, height difference, orientation difference, and volume difference. The building is then modeled using the fourth-order matrix to generate a fourth-order matrix model. A supervised clustering algorithm is used to train the fourth-order matrix model of the building, resulting in a fourth-order matrix intelligent clustering model. A fourth-order matrix intelligent clustering model is used to classify building morphology and identify landmark buildings, shanty buildings, and buildings with insufficient daylight. Combined with the coordinates of the center point of the target building plane, the diversity of the fourth-order matrix of buildings in a certain zone in the area is used as the spatial entropy measurement indicator of the zone. Urban areas are divided based on the lowest spatial entropy value of the zone, and morphologically chaotic areas, morphologically monotonous areas, and change fault zones in the city are identified. The object building classification and urban area zoning results are displayed in the form of point clouds of different colors. Multiple rounds of cyclic optimization are performed to display the optimized fourth-order matrix information of the object building, and the height, volume, orientation and coordinate information of the object building after modifying the fourth-order matrix information are obtained.

2. The automatic urban form measurement method based on the spatial entropy fourth-order matrix model according to claim 1 is characterized in that: The process of numbering the target buildings in the urban area, measuring and calculating the four morphological index data of the plane distance, height difference, orientation difference, and volume difference between each target building and the surrounding buildings by a total station, and obtaining the coordinate information of the plane center point of the target building by a GPS measuring instrument includes the following steps: Each building in the urban area is assigned a 14-digit code based on the following principles: the first six digits are the postal code of the administrative district where the building is located, the middle four digits are the block number, and the last four digits are the building number within the block; Using a total station, the relative data of the plane distance, height difference, orientation difference and volume difference between the target building and the surrounding buildings are obtained one by one according to the coding sequence; Using a GPS measuring instrument, the coordinates of the center points of the building plane are obtained one by one according to the coding sequence, and the center point coordinates are exported as ASCII coding format.

3. The automatic urban form measurement method based on the spatial entropy fourth-order matrix model according to claim 2 is characterized in that: The plane distance between the target building and the surrounding buildings was measured using the precision measurement mode of the total station with a minimum display unit of 1mm; the vertical angle difference between the target building and the surrounding buildings was measured using the horizontal angle measurement mode of the total station with a minimum angle measurement accuracy of 0.5°, and the height difference was obtained by combining the trigonometric height method; the orientation difference between the target building and the surrounding buildings was measured using the azimuth measurement mode of the total station with a minimum angle measurement accuracy of 0.5°; the volume difference between the target building and the surrounding buildings was obtained in the earthwork measurement mode of the total station, and the relative data of the four forms were exported in DAT format.

4. The automatic urban form measurement method based on the spatial entropy fourth-order matrix model according to claim 1 is characterized in that: The process of obtaining the fourth-order matrix intelligent clustering model includes the following steps: Using the obtained four morphological index data of plane distance, height difference, orientation difference and volume difference, a fourth-order matrix model is generated for each target building in the urban area. The four morphological index data are then linked to the fourth-order matrix model for display. The first row of the fourth-order matrix describes the plane distance between the target building and the surrounding buildings, the second row of the fourth-order matrix describes the height difference between the target building and the surrounding buildings, the third row of the fourth-order matrix describes the orientation difference between the target building and the surrounding buildings, and the fourth row of the fourth-order matrix describes the volume difference between the target building and the surrounding buildings. The four morphological index data of the fourth-order matrix model are used as machine learning labels. The target fourth-order matrix is divided into a training set, a validation set, and a test set in a ratio of 98:1:

1. A holographic sand table is used to demonstrate the fourth-order matrix of all buildings and building links in the urban area. Similar buildings of each building are found one by one on the holographic sand table, and the connection information is recorded and stored. A supervised clustering learning algorithm is used to perform fourth-order matrix clustering machine learning training through a graphics memory deep learning system. A machine learning model with strong generalization performance is selected as the fourth-order matrix automatic clustering model through cross-validation and generalization test.

5. The automatic urban form measurement method based on the spatial entropy fourth-order matrix model according to claim 4 is characterized in that: The number of columns of the fourth-order matrix is determined by the number of surrounding buildings of the target building.

6. The automatic urban form measurement method based on the spatial entropy fourth-order matrix model according to claim 1 is characterized in that: The process of classifying the morphology of the object building using the fourth-order matrix intelligent clustering model includes the following steps: The fourth-order matrix information of the target buildings in the urban area is classified using a fourth-order matrix automatic clustering model consisting of the distance difference, height difference, orientation difference and volume difference between the target building and the surrounding buildings; The fourth-order matrix information classified by the fourth-order matrix automatic clustering model is tested, among which the fourth-order matrix plane distance difference information value, height difference information value and volume difference information value of landmark buildings should be more than 20% higher than the average of other buildings; the fourth-order matrix plane distance difference information value and height difference information value of shanty buildings should be more than 20% lower than the average of other buildings; the fourth-order matrix plane distance difference information value and height difference information value of buildings with insufficient lighting should be more than 20% lower than the average of other buildings; when errors occur in the identification test results, the process of fourth-order matrix clustering machine learning training is repeated through the video memory deep learning system until the building information results classified by the automatic clustering model meet the test and identification requirements.

7. The automatic urban form measurement method based on the spatial entropy fourth-order matrix model according to claim 1 is characterized in that: The process of partitioning the urban area by combining the coordinates of the center point of the target building plane, taking the diversity of the fourth-order building matrix of a certain partition in the area as the spatial entropy measurement index of the partition, and taking the minimum spatial entropy value of the partition as the standard comprises the following steps: Cluster analysis is performed based on the coordinates of the center points of the building planes. Buildings with similar plane coordinates are divided into a zone, and the diversity of the fourth-order matrix changes in the zone is used as the measurement standard of spatial entropy. The measurement standard threshold of spatial entropy is set at 20%. If it is higher than the measurement standard threshold of spatial entropy, the morphological zone is reduced to 0.8 times of the original zone based on the classification results of the fourth-order matrix information of the building. Object buildings of a smaller number of classification types in the zone are eliminated, and then the zone is re-zoned. The morphological zone is compared with the spatial entropy measurement standard threshold again until the spatial entropy measurement standard range requirements are met. The regional morphological zone is optimized and the final regional morphological zone result is obtained. The calculation formula of spatial entropy is: H(x)=-∑p(xi)logp(xi) Where H(x) is the spatial entropy of partition x, and p(xi) is the ratio of the number of building type i in partition x to the total number of buildings in the partition. Identify morphologically chaotic areas, morphologically monotonous areas and change fault zones in urban areas. The basis for identifying morphologically chaotic areas is that the spatial entropy measurement value in the area is more than 3 times higher than that in other areas; the basis for identifying morphologically monotonous areas is that the spatial entropy measurement value in the area is less than 30% of that in other areas; and the change fault zone is that the fluctuation range of the spatial entropy measurement value of the adjacent surrounding areas is greater than 50%.

8. The automatic urban form measurement method based on the spatial entropy fourth-order matrix model according to claim 1 is characterized in that: The process of displaying the object building classification and urban area zoning results in the form of point clouds of different colors, performing multiple rounds of cyclic optimization, displaying the optimized fourth-order matrix information of the object building, and obtaining the height, volume, orientation, and coordinate information of the object building after modifying the fourth-order matrix information includes the following steps: The results of building classification and urban area zoning are displayed in the form of point clouds of different colors. The fourth-order matrix information of some buildings is modified based on six judgment rules: retaining landmark buildings, reducing shanty buildings, buildings with insufficient daylight, areas with chaotic morphology, areas with monotonous morphology, and changing fault zones. Optimize the fourth-order matrix information of the target building in a loop until the calculated values of the six types of judgment rules no longer change. Then output the modified fourth-order matrix information and classification and partitioning results, and display them in the holographic sandbox. Based on the modified fourth-order matrix information, the updated building height, volume, orientation and coordinate information are derived.

9. A device, characterized in that include: one or more processors; a memory for storing one or more programs; When one or more of the programs are executed by one or more of the processors, the one or more processors implement the automatic urban form measurement method based on the spatial entropy fourth-order matrix model as described in any one of claims 1-8.

10. A storage medium containing computer-executable instructions, characterized in that: When executed by a computer processor, the computer executable instructions are used to execute the urban form automatic measurement method based on the spatial entropy fourth-order matrix model as described in any one of claims 1 to 8.

Images