Unmanned aerial vehicle measurement system for building top surface

Abstract

The invention relates to the field of building measurement equipment, and discloses an unmanned aerial vehicle measurement system for a building top surface. The data acquisition module is used for acquiring image data and point cloud data of the top surface of the building and flight information of the unmanned aerial vehicle main body; the identification marking module is used for marking a building top surface structure according to the image data of the building top surface and preset structure data information; the calculation module is used for judging whether the marked image data of the top surface of the building is qualified or not according to a preset image precision threshold value, and calculating spatial feature data of the top surface of the building and generating a data re-collection instruction according to a qualified judgment result of the marked image data of the top surface of the building; and the path planning module is used for planning the flight path of the unmanned aerial vehicle according to the measurement requirement, the measurement area, the flight information and the data re-collection instruction. The method can solve the problem that it is difficult to accurately recognize the building top structure in the prior art.

Description

Technical Field

[0001] The present invention relates to the field of building measurement equipment, and in particular to a drone measurement system for building roofs. Background Art

[0002] Currently, the construction industry is rapidly developing, and architectural designs are becoming increasingly complex and diverse. As a crucial component of building structures, the condition of building roofs is directly related to the overall safety and performance of the building. Modern architecture places increasing demands on measurement accuracy and efficiency, making it crucial to accurately capture various parameters of building roofs. This is particularly important for measuring the size and location of roof openings such as ventilation openings, skylights, and manholes, as well as the size and location of protruding features such as beams and columns. This information is crucial for building equipment installation and spatial layout planning.

[0003] However, existing drone-based building roof measurement equipment has significant shortcomings. In terms of measurement accuracy, interference from external factors such as airflow and magnetic fields can lead to significant deviations in measurement data, which can easily lead to misjudgments of the actual building roof surface, seriously impacting subsequent building maintenance and repair efforts. In terms of data collection range, the equipment's sensor field of view is limited, making it difficult to fully capture data on complex building roof shapes. The omission of information on key areas hinders building space layout planning and poses potential risks to building safety assessments.

[0004] In view of this, it is necessary to design a new type of drone measurement equipment for building roofs to solve the problems of low measurement accuracy, incomplete data collection range, and difficulty in accurately identifying building roof structures in the existing technology. Summary of the Invention

[0005] In view of this, the present invention proposes a drone measurement system for building roofs, aiming to solve the problems in the existing technology of low measurement accuracy, incomplete data collection range, and difficulty in accurately identifying building roof structures.

[0006] In one aspect, the present invention provides a drone measurement system for building roofs, comprising:

[0007] UAV body;

[0008] Data acquisition module, used to collect image data of the building roof, point cloud data and flight information of the drone body;

[0009] an identification and marking module connected to the data acquisition module, the identification and marking module being used to receive the image data of the building top surface, the identification and marking module being used to mark the building top surface structure according to the image data of the building top surface and preset structure data information; and being used to identify all structures on the building top surface based on the marked building top surface structure, and output the marked building top surface image data;

[0010] a calculation module, connected to the data acquisition module and the identification and marking module, respectively, and configured to receive the point cloud data of the building top surface and the marked image data of the building top surface; the calculation module is further configured to determine whether the marked image data of the building top surface is qualified according to a preset image accuracy threshold, and calculate the spatial feature data of the building top surface and generate a data re-sampling instruction based on the qualified judgment result of the marked image data of the building top surface;

[0011] a path planning module, connected to the identification and marking module and the calculation module, respectively, for dividing the measurement area according to the image data of the building top surface; and for planning the flight path of the UAV according to the measurement requirements, the measurement area, flight information, and data re-collection instructions;

[0012] The data acquisition module is arranged at the bottom of the drone body, the identification mark module and the calculation module are both arranged inside the drone body, and the path planning module is arranged at the top of the drone body.

[0013] Furthermore, the calculation module calculates the spatial feature data of the building top surface and generates a data re-sampling instruction according to the qualified condition of the marked image data of the building top surface, including:

[0014] When it is determined that the marked image data of the building top surface is qualified, the calculation module calculates the building top surface spatial feature data based on the marked image data of the building top surface and the point cloud data;

[0015] When it is determined that the marked image data of the building top surface is unqualified, the calculation module generates a data re-collection instruction and transmits the data re-collection instruction to the data collection module and the path planning module.

[0016] Furthermore, the data acquisition module includes:

[0017] The sensor unit is used to collect data on the distance and position between structures on the roof of the building and transmit the data to the identification module. It is also used to collect information on the flight attitude, position and flight parameters of the drone and transmit the information to the calculation module and the path planning module.

[0018] An image acquisition unit, configured to acquire image data of a building top surface and transmit the image data to a recognition and marking module, and further configured to determine shape information of the building top surface based on the image data and transmit the shape information to a path planning module;

[0019] The laser radar unit is used to collect point cloud data of the building roof structure and transmit the point cloud data to the calculation module.

[0020] Furthermore, the identification mark module includes:

[0021] an image recognition unit, configured to pre-process the image data of the building top surface, extract features related to the structure from the pre-processed image data of the building top surface, perform feature matching between the extracted features related to the structure and preset structure data information, and identify the structure on the building top surface based on the matching results; the pre-processing includes gray-scaling, denoising, enhancing, and correcting the image data;

[0022] The marking unit is used to mark all the identified structures on the building top surface and transmit the marked image data of the building top surface to the calculation module.

[0023] Furthermore, the calculation module includes:

[0024] A data processing unit, configured to perform denoising and coordinate correction on the point cloud data;

[0025] A calculation unit is used to calculate and output the spatial feature data of the building top surface based on the marked image data of the building top surface and the point cloud data.

[0026] Furthermore, the system further comprises a judgment module, the judgment module being connected to the calculation module and configured to receive the building top surface spatial feature data output by the calculation module;

[0027] The judgment module has a built-in deep learning model, which judges whether the spatial feature data of the building top surface is reasonable based on a preset building top surface structure model, and determines the installation area of the marked structure based on the judgment result.

[0028] Furthermore, it also includes an anti-interference module, which is connected to the drone body and the data acquisition module respectively. The anti-interference module has a built-in magnetic compass, a GPS signal enhancer and a shock absorber.

[0029] Furthermore, it also includes a ground control terminal, which is used to input measurement task information and control the flight of the UAV measurement equipment, and is also used to receive and display the transmission data, recognition results, judgment results and flight path of each module in real time.

[0030] Furthermore, the process of the path planning module for dividing the measurement area according to the data information includes:

[0031] The path planning module receives the shape information transmitted by the image acquisition unit and performs preliminary segmentation of the shape information according to regular shapes;

[0032] The path planning module divides the rectangular shape surface in the shape information into equally spaced parallel lines;

[0033] The path planning module divides the irregular-shaped surface in the shape information into a plurality of approximate polygonal areas through a contour recognition algorithm.

[0034] Furthermore, the roof structure of the building includes: chimneys, vents, skylights, manholes, beams, columns, elevator shafts and photovoltaic panels;

[0035] The building top surface spatial characteristic data includes: data on the size, position and area of the building top surface structure and area data of the building top surface excluding the structure.

[0036] Furthermore, data exchange is performed between the data acquisition module, identification and marking module, calculation module, judgment module, anti-interference module, path planning module and ground control terminal via wired or wireless means.

[0037] Compared with the prior art, the present invention has the following beneficial effects:

[0038] This application simultaneously collects multi-source information such as image data, point cloud data, and flight information of the building roof, providing a more comprehensive data basis for subsequent analysis and judgment, which can more accurately describe the characteristics of the building roof and improve the comprehensiveness and accuracy of the measurement data.

[0039] This application uses a recognition and marking module to accurately identify and mark various structures on building roofs. Compared to traditional methods that rely on manual recognition or simple image analysis, this application significantly improves recognition efficiency and accuracy, reducing human error. Furthermore, this application can quickly output marked image data, providing a clear data foundation for subsequent calculations and analysis.

[0040] By implementing a calculation module, this application not only calculates the spatial feature data of building roof surfaces based on point cloud data and image data, but also determines the eligibility of image data based on a preset image accuracy threshold. If image data fails to meet the requirements, a data re-collection instruction is generated promptly, ensuring the quality of the measurement data. This intelligent calculation and decision-making mechanism avoids erroneous analysis and decision-making due to data quality issues, improving the stability and reliability of the entire measurement system.

[0041] Unlike traditional fixed paths or simple obstacle avoidance path planning, the path planning module of this application can more flexibly adapt to the shapes and measurement requirements of different building roofs, achieve more efficient and comprehensive measurement coverage, reduce drone flight time and energy consumption, and improve measurement efficiency.

[0042] The modules of this application are tightly integrated and work together to form an organic whole, making the entire measurement process smoother and more efficient, reducing information transmission loss and errors between modules, and improving the overall performance of the system.

[0043] Compared with the existing technology, the present invention can obtain building top surface data more comprehensively and accurately and identify building top surface structures. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Various other advantages and benefits will become apparent to those skilled in the art upon reading the detailed description of the preferred embodiment below. The accompanying drawings are for illustration purposes only and are not to be considered as limiting the present invention. The same reference symbols are used throughout the drawings to represent the same components. In the drawings:

[0045] Figure 1 This is a functional block diagram of a drone measurement system for building roofs provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0046] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art. It should be noted that, unless there is a conflict, the embodiments of the present disclosure and the features in the embodiments can be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in conjunction with the embodiments.

[0047] Currently, the construction industry is rapidly developing, and architectural designs are becoming increasingly complex and diverse. As a crucial component of building structures, the condition of building roofs is directly related to the overall safety and performance of the building. Modern architecture places increasing demands on measurement accuracy and efficiency, making it crucial to accurately capture various parameters of building roofs. This is particularly important for measuring the size and location of roof openings such as ventilation openings, skylights, and manholes, as well as the size and location of protruding features such as beams and columns. This information is crucial for building equipment installation and spatial layout planning.

[0048] However, existing drone-based building roof measurement equipment has significant shortcomings. In terms of measurement accuracy, interference from external factors such as airflow and magnetic fields can lead to significant deviations in measurement data, which can easily lead to misjudgments of the actual building roof surface, seriously impacting subsequent building maintenance and repair efforts. In terms of data collection range, the equipment's sensor field of view is limited, making it difficult to fully capture data on complex building roof shapes. The omission of information on key areas hinders building space layout planning and poses potential risks to building safety assessments.

[0049] Therefore, the present invention proposes a drone measurement system for building roofs to solve the problems of low measurement accuracy and incomplete data collection range in the existing technology, and can accurately and effectively identify the structures on the building roofs.

[0050] Reference Figure 1 As shown, in some embodiments of the present application, a drone measurement system for a building roof includes: a drone body, a data acquisition module, an identification and marking module, a calculation module, and a path planning module.

[0051] Specifically, the data acquisition module is used to collect image data of the building roof, point cloud data and flight information of the drone body.

[0052] Specifically, the flight information of the drone includes the drone's flight attitude, position, and flight parameters. Flight parameters include the drone's maximum turning radius and minimum flight altitude.

[0053] Specifically, the identification and marking module is connected to the data acquisition module and is used to receive the image data of the building top surface. The identification and marking module is used to mark the building top surface structure according to the image data of the building top surface and the preset structure data information; it is also used to identify all structures on the building top surface based on the marked building top surface structure, and output the marked building top surface image data.

[0054] Specifically, the identification and marking module uses image processing algorithms to extract structure-related features from image data of building roofs. Common features include edge features, texture features, and color features. Edge detection algorithms are used to extract the outlines of structures on building roofs; texture analysis algorithms are used to analyze the surface texture information; and color distribution features are used to identify structures of specific colors.

[0055] Specifically, the preset structure data information includes the geometric shape characteristics of the structure, such as rectangle, circle, triangle, etc.; structural characteristics, such as frame structure, solid structure, etc.; functional characteristics, such as chimney vents, etc.; and material characteristics.

[0056] It is understood that the recognition and marking module matches features extracted from the image data of the building roof with features in the preset structure data information. This can be done using methods such as template matching and feature point matching. For example, a rectangular outline detected in the image is matched with the rectangular shape features of a vent on the building roof in the preset structure data information. If a match is successful, the rectangular outline is marked as a vent.

[0057] It can be understood that, through the above means, the identification and marking module can identify and mark all structures on the top surface of the building.

[0058] Specifically, the calculation module is connected to the data acquisition module and the identification and marking module respectively, and the calculation module is used to receive the point cloud data of the building top surface and the image data of the marked building top surface; the calculation module is also used to judge whether the marked image data of the building top surface is qualified according to a preset image accuracy threshold, and calculate the spatial feature data of the building top surface and generate data re-acquisition instructions based on the qualified judgment result of the marked image data of the building top surface.

[0059] Specifically, the calculation module compares the marked image data of the building top surface with a preset image accuracy threshold. If the accuracy of the marked image data of the building top surface reaches the preset image accuracy threshold, the calculation module determines that the marked image data of the building top surface is qualified. If the accuracy of the marked image data of the building top surface does not reach the preset image accuracy threshold, the calculation module determines that the marked image data of the building top surface is unqualified and generates a data re-collection instruction.

[0060] Specifically, image accuracy includes image spatial resolution, image geometric accuracy, and image classification accuracy.

[0061] Specifically, the path planning module is connected to the identification and marking module and the calculation module respectively. The path planning module is used to divide the measurement area according to the image data of the building roof; and is also used to plan the UAV flight path according to the measurement requirements, the measurement area, flight information and data re-collection instructions.

[0062] Specifically, the data acquisition module is arranged at the bottom of the drone body, the identification mark module and the calculation module are both arranged inside the drone body, and the path planning module is arranged at the top of the drone body.

[0063] Specifically, the data acquisition module is arranged at the bottom of the drone body, the identification mark module and the calculation module are both arranged inside the drone body, and the path planning module is arranged at the top of the drone body.

[0064] It can be seen that the modules of this application are tightly integrated and work together to form an organic whole, making the entire measurement process smoother and more efficient, reducing information transmission loss and errors between modules, and improving the overall performance of the system.

[0065] Reference Figure 1 As shown, in some embodiments of the present application, the process of the calculation module calculating the building top surface spatial feature data and generating the data re-sampling instruction according to the qualified status of the marked image data of the building top surface includes:

[0066] When it is determined that the image data of the building top surface after marking is qualified, the calculation module calculates the spatial feature data of the building top surface based on the image data of the building top surface after marking and the point cloud data; when it is determined that the image data of the building top surface after marking is unqualified, the calculation module generates a data re-sampling instruction and transmits the data re-sampling instruction to the data acquisition module and the path planning module.

[0067] Specifically, before calculating the spatial feature data of the building roof, the calculation module first performs operations such as denoising and contrast enhancement on the marked image data of the building roof to extract the contour features of the image data. It then filters the point cloud data to remove outliers and noise points and extract the range features of the point cloud data. The calculation module then matches the contour features of the image data with the range features of the point cloud data to ensure that they accurately correspond in the spatial coordinate system.

[0068] Specifically, the calculation module calculates the spatial feature data of the building roof based on the contour features of the image data and the range features of the point cloud data, such as the basic dimensional information such as the length, width, and area of the building roof, as well as the dimensional information such as the length, width, and height of the structures within the building roof.

[0069] It is understandable that the calculation module fuses the spatial feature data calculated from the image data and the point cloud data, complements and verifies each other, so as to improve the accuracy of the measurement.

[0070] It is understood that the calculation module can perform error analysis and quality assessment on the calculated spatial feature data to check the accuracy and reliability of the data. If large errors or unreasonable situations are found in the data, it is necessary to recalculate and make necessary adjustments and corrections.

[0071] Reference Figure 1 As shown, in some embodiments of the present application, the data acquisition module includes a sensor unit, an image acquisition unit and a lidar unit.

[0072] Specifically, the sensor unit is used to collect data on the distance and position between structures on the building's rooftops and transmit this data to the identification and marking module. It is also used to collect information on the drone's flight attitude, position, and flight parameters and transmit this information to the calculation module and path planning module. The image acquisition unit is used to collect image data of the building's rooftops and transmit this image data to the identification and marking module. It is also used to determine the shape of the building's rooftops based on this image data and transmit this shape information to the path planning module. The lidar unit is used to collect point cloud data of structures on the building's rooftops and transmit this point cloud data to the calculation module.

[0073] Reference Figure 1 As shown, in some embodiments of the present application, the identification and marking module includes an image recognition unit and a marking unit.

[0074] Specifically, it is used to preprocess the image data of the building top surface and extract features related to the structure from the preprocessed image data of the building top surface. It is also used to match the extracted features related to the structure with preset structure data information, and identify the structure on the building top surface based on the matching results.

[0075] It is understood that the image recognition unit performs feature matching on the preprocessed image data with pre-set data information on structures such as chimneys and skylights. For example, it compares the contour features and texture features in the image with corresponding features in the pre-set data to identify the structures on the building roof. Based on the recognition results, the marking unit marks all identified structures on the building roof with different colored borders or symbols, such as a red border for chimneys and a green border for vents, and transmits the marked image data to the computing module.

[0076] Specifically, the preprocessing includes: graying, denoising, enhancing and correcting the image data of the building top surface.

[0077] Specifically, the marking unit is used to mark all the identified structures on the building top surface, and transmit the marked image data of the building top surface to the calculation module.

[0078] It can be seen that during the measurement process, the data acquisition module continuously transmits the collected data on the distance and position between the structures on the top of the building and the image data of the top of the building to the identification and marking module. The identification and marking module preprocesses these data to help improve the accuracy of the measurement data.

[0079] Reference Figure 1 As shown, in some embodiments of the present application, the computing module includes a data processing unit and a computing unit.

[0080] Specifically, the data processing unit is used to perform denoising and coordinate correction on the point cloud data; the calculation unit is used to calculate and output the spatial feature data of the building top surface based on the marked image data of the building top surface and the point cloud data.

[0081] It is understood that after receiving the point cloud data and labeled image data of the building's roof, the data processing unit denoises the received point cloud data, removing outliers caused by LiDAR measurement errors or external interference, and using algorithms such as bilateral filtering to preserve the detailed features of the point cloud data. At the same time, the computing module corrects the coordinates of the original point cloud data based on the drone's flight attitude and position to ensure the accuracy of the spatial position of the point cloud data. Based on the labeled image data and processed point cloud data, the computing unit uses spatial geometry algorithms to calculate the spatial feature data of the building's roof, such as the actual area of the building's roof, the floor space and volume of various structures, and outputs this data for subsequent analysis.

[0082] Reference Figure 1 As shown, in some embodiments of the present application, a judgment module is also included.

[0083] Specifically, the judgment module is connected to the calculation module, and the judgment module is used to receive the building top surface space feature data output by the calculation module.

[0084] Specifically, the judgment module has a built-in deep learning model, which judges whether the spatial feature data of the building top surface is reasonable based on a preset building top surface structure model, and determines the installation area of the marked structure based on the judgment result.

[0085] It is understandable that the calculation module outputs the building top surface space feature data, such as the size, location and area of various structures, to the judgment module. The judgment module has a built-in neural network model based on deep learning, which compares and analyzes the received building top surface space feature data with the preset building top surface structure model. For example, according to the preset model, it is judged whether the layout of chimneys and vents in a certain area is reasonable, and whether the sizes of various structures meet the design standards. The installation area of the marked structure is determined based on the judgment results. If it is found that the vents in a certain area are too dense and do not meet the ventilation efficiency requirements, the area is judged as an area where new vents cannot be installed, and these analysis results are fed back to relevant personnel to provide a reference for subsequent building planning and maintenance.

[0086] Reference Figure 1 As shown, in some embodiments of the present application, an anti-interference module is also included, which is connected to the drone body and the data acquisition module respectively. The anti-interference module has a built-in magnetic compass, a GPS signal enhancer and a shock absorber.

[0087] It is understood that the magnetic compass monitors the drone's heading in real time. When external electromagnetic interference causes abnormal heading data, the magnetic compass can promptly adjust and stabilize the heading. The GPS signal booster enhances the GPS signal strength received by the drone, ensuring that even in weak signal environments such as those obstructed by tall buildings, the drone can still accurately obtain its own position information and maintain the accuracy of its flight path. The shock absorber is installed at the connection between the drone's data acquisition module and the fuselage to reduce the impact of vibrations caused by air turbulence and other factors on the data acquisition equipment during flight. This ensures that the sensor unit, image acquisition unit, and lidar unit can collect data stably, improving the accuracy and reliability of data acquisition.

[0088] Reference Figure 1 As shown, in some embodiments of the present application, a ground control terminal is also included.

[0089] Specifically, the ground control terminal is used to input measurement task information and control the flight of the UAV measurement equipment, and is also used to receive and display the transmission data, recognition results, judgment results and flight path of each module in real time.

[0090] As you can understand, the ground control terminal controls the takeoff of the drone measurement equipment and monitors its flight status in real time. The ground control terminal receives real-time information, including building roof image data and point cloud data from the data acquisition module, recognition results from the identification and marking module, calculation results from the calculation module, and the flight path planned by the path planning module, and displays it in an intuitive graphical interface. Based on this real-time information, the operator can adjust the measurement strategy promptly. If poor image quality is detected in a certain area, the ground control terminal can issue a command to re-collect data in that area.

[0091] Specifically, data is exchanged between the data acquisition module, the identification and marking module, the calculation module, the judgment module, the anti-interference module, the path planning module and the ground control terminal via wired or wireless means.

[0092] Reference Figure 1 As shown, in some embodiments of the present application, the process of the path planning module for dividing the measurement area according to the data information includes:

[0093] The path planning module receives the shape information transmitted by the image acquisition unit and preliminarily segments the shape information according to regular shapes; the path planning module divides the rectangular shape surface in the shape information according to equally spaced parallel lines; the path planning module segments the irregular shape surface in the shape information into multiple approximate polygonal areas through a contour recognition algorithm.

[0094] It is understandable that when the path planning module processes the image data of a building roof with an irregular shape, it first receives the shape information from the image acquisition unit and finds that the overall shape of the building roof is relatively complex, including rectangular areas and irregular polygonal areas. The path planning module preliminarily divides the overall shape information according to regular shapes and identifies the rectangular areas therein. For rectangular surfaces, the path planning module divides them according to equally spaced parallel lines. For example, the parallel line spacing is set to 5 meters, and the rectangular area is divided into multiple long strip measurement areas with a width of 5 meters. For irregular surfaces in the shape information, the path planning module uses contour recognition algorithms, such as the Canny edge detection algorithm combined with the polygon approximation algorithm, to divide them into multiple approximate polygonal areas, such as dividing an irregular roof garden area into several triangular and quadrilateral areas, so that the drone can collect data from these areas more efficiently.

[0095] Reference Figure 1 As shown, in some embodiments of the present application, the building roof structure includes: chimneys, vents, skylights, manholes and other openings, beams, columns and other protrusions, as well as elevator shafts and photovoltaic panels.

[0096] The building top surface spatial characteristic data includes: data on the size, position and area of the building top surface structure and area data of the building top surface excluding the structure.

[0097] The drone measurement system for building roofs in the above-mentioned embodiment simultaneously collects multi-source information such as image data, point cloud data, and flight information of the building roofs, providing a more comprehensive data basis for subsequent analysis and judgment, and can more accurately describe the characteristics of the building roofs, thereby improving the comprehensiveness and accuracy of the measurement data.

[0098] As can be seen, this application can accurately identify and mark various structures on the roof of a building through the identification and marking module. Compared with traditional methods that rely on manual recognition or simple image analysis, this application greatly improves recognition efficiency and accuracy and reduces human error. At the same time, this application can quickly output marked image data, providing a clear data foundation for subsequent calculations and analysis.

[0099] As can be seen, by setting up a calculation module, this application can not only calculate the spatial feature data of the building roof based on point cloud data and image data, but also determine whether the image data is qualified based on a preset image accuracy threshold. When the image data is unqualified, a data re-collection instruction is generated in a timely manner to ensure the quality of the measurement data. This intelligent calculation and decision-making mechanism of the present application avoids erroneous analysis and decision-making caused by data quality issues, and improves the stability and reliability of the entire measurement system.

[0100] It can be seen that, unlike traditional fixed paths or simple obstacle avoidance path planning, the path planning module of this application can more flexibly adapt to the shapes and measurement requirements of different building roofs, achieve more efficient and comprehensive measurement coverage, reduce drone flight time and energy consumption, and improve measurement efficiency.

[0101] It can be seen that the modules of this application are tightly integrated and work together to form an organic whole, making the entire measurement process smoother and more efficient, reducing information transmission loss and errors between modules, and improving the overall performance of the system.

[0102] It can be seen that compared with the existing technology, the present invention can obtain building top surface data more comprehensively and accurately and identify building top surface structures.

[0103] 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 take the form of a complete hardware embodiment, a complete software embodiment, or a combination of software and hardware embodiments. Furthermore, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0104] 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 produce 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 gardening device having the functions specified in one or more boxes.

[0105] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, such that the instructions stored in the computer readable memory produce an article of manufacture comprising a gardening device that implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0106] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0107] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered by the scope of protection of the claims of the present invention.

Claims

1. A drone measurement system for building roofs, characterized in that: include: UAV body; Data acquisition module, used to collect image data of the building roof, point cloud data and flight information of the drone body; an identification and marking module connected to the data acquisition module, the identification and marking module being used to receive the image data of the building top surface, and the identification and marking module being used to mark the building top surface structure according to the image data of the building top surface and preset structure data information; It is also used to identify all structures on the building top surface according to the marked building top surface structures, and output the image data of the marked building top surface; a calculation module, connected to the data acquisition module and the identification and marking module, respectively, and configured to receive the point cloud data of the building top surface and the marked image data of the building top surface; the calculation module is further configured to determine whether the marked image data of the building top surface is qualified according to a preset image accuracy threshold, and calculate the spatial feature data of the building top surface and generate a data re-sampling instruction based on the qualified judgment result of the marked image data of the building top surface; a path planning module, connected to the identification and marking module and the calculation module, respectively, for dividing the measurement area according to the image data of the building top surface; and for planning the flight path of the UAV according to the measurement requirements, the measurement area, flight information, and data re-collection instructions; The data acquisition module is arranged at the bottom of the drone body, the identification mark module and the calculation module are both arranged inside the drone body, and the path planning module is arranged at the top of the drone body.

2. The drone measurement system for building roof according to claim 1, characterized in that: The process of the calculation module calculating the spatial feature data of the building top surface and generating a data re-sampling instruction according to the qualified condition of the marked image data of the building top surface includes: When it is determined that the marked image data of the building top surface is qualified, the calculation module calculates the building top surface spatial feature data based on the marked image data of the building top surface and the point cloud data; When it is determined that the marked image data of the building top surface is unqualified, the calculation module generates a data re-collection instruction and transmits the data re-collection instruction to the data collection module and the path planning module.

3. The drone measurement system for building roof according to claim 1, characterized in that: The data acquisition module includes: The sensor unit is used to collect data on the distance and position between structures on the roof of the building and transmit the data to the identification module. It is also used to collect information on the flight attitude, position and flight parameters of the drone and transmit the information to the calculation module and the path planning module. An image acquisition unit, configured to acquire image data of a building top surface and transmit the image data to a recognition and marking module, and further configured to determine shape information of the building top surface based on the image data and transmit the shape information to a path planning module; The laser radar unit is used to collect point cloud data of the building roof structure and transmit the point cloud data to the calculation module.

4. The drone measurement system for building roofs according to claim 3, characterized in that: The identification mark module includes: an image recognition unit, configured to pre-process the image data of the building top surface, extract features related to the structure from the pre-processed image data of the building top surface, perform feature matching between the extracted features related to the structure and preset structure data information, and identify the structure on the building top surface based on the matching results; the pre-processing includes gray-scaling, denoising, enhancing, and correcting the image data; The marking unit is used to mark all the identified structures on the building top surface and transmit the marked image data of the building top surface to the calculation module.

5. The drone measurement system for building roof according to claim 1, characterized in that: The calculation module includes: A data processing unit, configured to perform denoising and coordinate correction on the point cloud data; A calculation unit is used to calculate and output the spatial feature data of the building top surface based on the marked image data of the building top surface and the point cloud data.

6. The drone measurement system for building roofs according to claim 1, characterized in that: The invention also includes a judgment module, the judgment module is connected to the calculation module, and the judgment module is used to receive the building top surface space feature data output by the calculation module; The judgment module has a built-in deep learning model, which judges whether the spatial feature data of the building top surface is reasonable based on a preset building top surface structure model, and determines the installation area of the marked structure based on the judgment result.

7. The drone measurement system for building roofs according to claim 1, characterized in that: It also includes an anti-interference module, which is connected to the drone body and the data acquisition module respectively. The anti-interference module has a built-in magnetic compass, a GPS signal enhancer and a shock absorber.

8. The drone measurement system for building roofs according to claim 1, characterized in that: It also includes a ground control terminal, which is used to input measurement task information and control the flight of the UAV measurement equipment, and is also used to receive and display the transmission data, recognition results, judgment results and flight path of each module in real time.

9. The drone measurement system for building roofs according to claim 3, characterized in that: The process of the path planning module for dividing the measurement area according to the data information includes: The path planning module receives the shape information transmitted by the image acquisition unit and performs preliminary segmentation of the shape information according to regular shapes; The path planning module divides the rectangular shape surface in the shape information into equally spaced parallel lines; The path planning module divides the irregular-shaped surface in the shape information into a plurality of approximate polygonal areas through a contour recognition algorithm.

10. The drone measurement system for building roofs according to claim 6, characterized in that: The roof structures of the building include: chimneys, vents, skylights, manholes, beams, columns, elevator shafts and photovoltaic panels; The building top surface spatial characteristic data includes: data on the size, position and area of the building top surface structure and area data of the building top surface excluding the structure.

Images