Real estate project quality detection method and system based on drone inspection

Abstract

The present application provides a real estate project quality inspection method and system based on drone inspection, which relates to the field of quality inspection technology. The method includes: obtaining a collection path by planning a drone inspection path; constructing data space coordinates; establishing a quality inspection category framework; inputting drone inspection data into the quality inspection category framework based on the data space coordinates to perform inspection path matching and import, and obtain quality inspection results for the corresponding quality inspection category. This application solves the technical problem that the accuracy and efficiency of real estate project quality inspection are affected by the relatively complex building structure and many interference factors of real estate projects, resulting in low drone inspection accuracy. By planning the inspection path, the complex building structure is fully covered, the drone inspection accuracy is improved, and the accuracy and efficiency of real estate project quality inspection in a high-interference environment are improved.

Description

Technical Field

[0001] The present application relates to the field of quality inspection technology, and in particular to a real estate project quality inspection method and system based on drone inspection. Background Art

[0002] Real estate project quality inspection is a crucial step in ensuring building safety, functionality, and durability. Traditional methods rely primarily on manual inspections. However, due to the complex structures and extensive areas of buildings, inspection efficiency is low, the workload is high, and the inspection is susceptible to human interference. Especially in high-rise buildings or complex terrain, manual inspections are difficult to fully guarantee accuracy and reliability.

[0003] With the rapid development of drone technology, its application has become increasingly widespread across multiple industries, particularly in the field of building quality inspection, where an increasing number of real estate projects are beginning to employ drones for inspections. However, due to the varying structural characteristics and quality issues faced by different building types, the inspection requirements and methods for each building vary. In some complex structures, drone inspection path planning and data collection may have blind spots. This is particularly true in hard-to-reach areas such as high-rise buildings, corners, and mezzanines, where drone inspection technology cannot cover all critical areas. Furthermore, environmental interference, equipment limitations, and data processing complexity during flight not only impact the accuracy and efficiency of drone inspections but also challenge the overall accuracy of quality inspections.

[0004] To sum up, there are technical problems in the existing technology, such as the low accuracy of drone detection due to the complex architectural structure of real estate projects and the large number of interference factors, which affects the accuracy and efficiency of real estate project quality inspection. Summary of the Invention

[0005] The purpose of this application is to provide a real estate project quality inspection method and system based on drone inspection, in order to solve the technical problem in the existing technology that the drone inspection accuracy is low due to the complex architectural structure of the real estate project and the large number of interference factors, thereby affecting the accuracy and efficiency of the real estate project quality inspection.

[0006] In view of the above problems, this application provides a real estate project quality inspection method and system based on drone inspection.

[0007] In the first aspect, the present application provides a real estate project quality inspection method based on drone inspection, and the real estate project quality inspection method based on drone inspection is implemented by a real estate project quality inspection system based on drone inspection, wherein the real estate project quality inspection method based on drone inspection includes: planning the drone inspection path according to the spatial distribution relationship of quality inspection, and obtaining the data source collection path; locating the collected data according to the data source collection path, and constructing data space coordinates; establishing a quality inspection category framework, and the quality inspection category framework includes quality inspection category, data requirement parameters, identification detection operator and detection path; based on the data space coordinates, the drone inspection data is input into the quality inspection category framework for detection path matching and import, the inspection data is identified and extracted through the data requirement parameters, and the detection category identification processing is performed using the identification detection operator to obtain the quality inspection result of the corresponding quality inspection category.

[0008] Optionally, a geographic three-dimensional model of the building to be inspected is constructed to identify the building structure; the building structure is decomposed according to the quality inspection requirements to determine the inspection space structure, and the inspection space structure has quality inspection requirement identification characteristics; the inspection space structure is inspected for spatial characteristics and spatial connection relationship identification to obtain the spatial distribution relationship of quality inspection.

[0009] Optionally, the inspection space characteristics include the exterior and interior of the building, the spatial connection relationship represents the inspection space structural relationship and spatial connection position range between the exterior and interior of the building, and the spatial distribution relationship of the quality inspection includes quality inspection requirements, inspection space characteristics and spatial relationships.

[0010] Optionally, according to the multi-source acquisition characteristics of the sensors carried by the drone and combined with historical record data, the detection space characteristics and acquisition data characteristics of the multi-source sensors are determined, wherein the multi-source sensors include cameras, infrared thermal imagers, lidars, and ultrasonic sensors; according to the detection space characteristics and acquisition data characteristics of the multi-source sensors, the detection requirements are matched with the spatial distribution relationship of the quality detection, and an acquisition mapping of the multi-source sensors and the spatial distribution relationship is established; according to the acquisition mapping combined with the acquisition space constraints of the multi-source sensors, path planning is performed with the goal of full coverage of the detection spatial distribution relationship and path minimization to obtain the data source acquisition path.

[0011] Optionally, according to the spatial distribution relationship of the quality detection, regional division is performed to obtain the drone collaborative collection area, wherein the regional division includes the division of the same plane space and the division of the opposite surfaces of the building space; the spatial detection constraints between the areas in the regional division results are obtained; according to the spatial detection constraints, the drone collection paths of each area are collaboratively planned to obtain the data source collection path, which includes multiple collaborative drone data source collection paths.

[0012] Optionally, according to the quality inspection requirements corresponding to the spatial inspection constraints, a quality inspection priority is obtained, and the quality inspection priority is proportional to the impact of quality defects and the timeliness of inspection; according to the quality inspection priority, the collaborative planning weights of the spatial inspection constraints are configured; based on the spatial detection constraints and their collaborative planning weights, the drone collection paths in each area are collaboratively planned.

[0013] Optionally, the structural position of the building to be inspected where the inspection data is located is located according to the data source acquisition path to obtain the building positioning coordinates of the inspection data; based on the spatial relationship of the building structure, a spatial coordinate system is constructed; the building positioning coordinates of all inspection data are projected into the spatial coordinate system to obtain the data spatial coordinates, wherein the inspection data has sensor type and quality detection requirements.

[0014] Optionally, based on the detection timeliness of each quality detection requirement, a detection timeliness hierarchy is established; based on the data requirement parameters of each quality detection requirement and the processing process of the identification detection operator, the processing relationship is fitted with the multi-source collected data of the drone to obtain the processing identification path of each quality detection requirement; based on the processing identification path of each quality detection requirement, the data transmission requirements and computing power requirements of each path node are obtained, and an operation requirement chain is established; with the detection time threshold of each quality detection requirement in the detection timeliness hierarchy as a constraint, the edge device nodes of the operation requirement chain are matched to construct the quality detection category framework, which is used to reflect the edge device node path and data requirement parameters and identification detection operators corresponding to the quality detection requirements during the processing process.

[0015] Optionally, data alignment and integration are performed according to the data space coordinates, sensor type, and quality inspection requirements of the inspection data; the integrated inspection data is input into the quality inspection category framework, the corresponding edge device nodes are matched, and quality inspection of the corresponding building structure position is performed according to the data space coordinates.

[0016] In the second aspect, the present application also provides a real estate project quality inspection system based on drone inspection, which is used to execute the real estate project quality inspection method based on drone inspection as described in the first aspect, wherein the real estate project quality inspection system based on drone inspection includes: a path planning module, which is used to plan the drone inspection path according to the spatial distribution relationship of quality inspection and obtain the data source collection path; a coordinate positioning module, which is used to locate the collected data according to the data source collection path and construct data space coordinates; a framework construction module, which is used to establish a quality inspection category framework, and the quality inspection category framework includes quality inspection category, data requirement parameters, identification detection operator and detection path; a detection result generation module, which is used to input the drone inspection data into the quality inspection category framework based on the data space coordinates for detection path matching and import, identify and extract the inspection data through the data requirement parameters, and use the identification detection operator to perform detection category identification processing to obtain quality inspection results corresponding to the quality inspection category.

[0017] One or more technical solutions provided in this application have at least the following beneficial effects:

[0018] The invention relates to a method for accurately locating the data source collection path based on the spatial distribution of quality inspections. The method locates the collected data according to the data source collection path and constructs data spatial coordinates. The method establishes a quality inspection category framework, which includes quality inspection categories, data requirement parameters, identification detection operators, and inspection paths. The method then inputs the drone inspection data into the quality inspection category framework based on the data spatial coordinates for inspection path matching and importing. The inspection data is identified and extracted using the data requirement parameters, and the identification detection operators are used to perform inspection category identification processing to obtain quality inspection results corresponding to the quality inspection categories. In other words, by analyzing the spatial distribution of quality inspections and performing path planning, the drone-collected data is accurately located with the building structure, reducing errors caused by interference factors and ensuring that each piece of data truly reflects the actual structural state. The method constructs a comprehensive and efficient quality inspection category framework. After the inspection data is input into the quality inspection framework, the detection operators are used to automatically identify and output quality inspection results. This overcomes the limitation of drones in detecting small or complex defects in the quality inspection of high-rise buildings and complex components, thereby improving the accuracy and efficiency of real estate project quality inspections.

[0019] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, which can be implemented in accordance with the contents of the description, and to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are specifically listed below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions in this application or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are merely exemplary, and a person of ordinary skill in the art can obtain other drawings based on the provided drawings without creative work.

[0021] Figure 1 This is a flow chart of the real estate project quality inspection method based on drone inspection in this application.

[0022] Figure 2 This is a structural diagram of the real estate project quality inspection system based on drone inspection in this application.

[0023] Description of the accompanying drawings: path planning module 11, coordinate positioning module 12, framework construction module 13, detection result generation module 14. DETAILED DESCRIPTION

[0024] This application provides a real estate project quality inspection method and system based on drone inspections, solving the technical problem in the prior art of low drone inspection accuracy due to the complex architectural structure and numerous interference factors of real estate projects, which affects the accuracy and efficiency of real estate project quality inspections. By analyzing the spatial distribution of quality inspections and performing path planning, the data collected by drones is accurately positioned with the building structure. A comprehensive and efficient quality inspection category framework is then constructed. After the inspection data is input into the quality inspection framework, the detection operator is used to automatically identify and output the quality inspection results. This overcomes the limitation of drones in detecting small or complex defects in the quality inspection of high-rise buildings and complex components, thereby improving the accuracy and efficiency of real estate project quality inspections.

[0025] Below, the technical solutions in this application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of this application, rather than all the embodiments of this application. It should be understood that this application is not limited to the example embodiments described herein. Based on the embodiments of this application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application. It should also be noted that, for the convenience of description, only the parts related to this application, rather than all of them, are shown in the accompanying drawings.

[0026] For example, see the attached Figure 1 The present application provides a real estate project quality inspection method based on drone inspection, wherein the real estate project quality inspection method based on drone inspection is performed by a real estate project quality inspection system based on drone inspection, and the real estate project quality inspection method based on drone inspection specifically includes the following steps:

[0027] S100: According to the spatial distribution relationship of quality inspection, the drone inspection path is planned to obtain the data source collection path.

[0028] Specifically, drone inspection routes are planned based on the spatial distribution relationship of quality inspections obtained as described below. The spatial distribution relationship of quality inspections refers to determining the spatial layout of various inspection areas in a building, as well as the connection relationships between them, based on different inspection requirements and building structures. Based on the spatial distribution relationship, specific inspection areas are divided, and the inspection tasks for each area are defined. Based on the spatial layout and inspection requirements of the building, route planning is carried out according to certain rules to ensure that inspections cover every important inspection area, while avoiding duplicate inspections and unnecessary route deviations. Drone inspection route planning must not only consider the layout of the building, but also the drone's flight capabilities (such as flight altitude, flight speed, and flight distance), environmental impacts (such as wind speed, obstacles, etc.), and the priority of inspection tasks.

[0029] For example, the A-star algorithm is used for drone path planning. The A-star algorithm selects the optimal path by calculating the estimated cost from each node to the target node (based on heuristic estimates) and the actual cost from the starting point to the current node. First, a three-dimensional building model with clear spatial distribution relationships is converted into a graph, with each inspection area treated as a node in the graph. The A-star algorithm analyzes the spatial relationships between nodes and possible paths to calculate the shortest path from the starting point to the destination, selecting the optimal path to avoid obstacles (such as building structures and other equipment).

[0030] During the inspection process, a drone flies along a planned inspection route, collecting various data (such as images, videos, temperature and humidity, infrared thermal imaging, and laser scanning point clouds). Each inspection route corresponds to a specific data collection task, which may include image acquisition, temperature monitoring, crack detection, humidity sensing, and other data types. The data collection route must be planned to ensure that the drone can collect the data required to cover all quality inspection areas during the inspection. For example, when inspecting a facade, the drone may need to capture high-definition images to detect cracks, coating peeling, and other issues. Furthermore, an infrared thermal imager may be required to detect temperature differences within the wall to identify thermal bridges and potential moisture issues. The collection route will guide the drone to different locations based on the inspection area, automatically activating sensors for data collection.

[0031] After a drone completes its inspection mission along a planned route, all generated data is recorded and analyzed within the quality inspection category framework. Through precise path planning, the drone can cover all key building areas, including the facade, roof, basement, and interior spaces, ensuring that nothing is missed. The path planning algorithm designs the shortest and safest inspection route, reducing drone flight time and energy consumption, and improving the accuracy, efficiency, and intelligence of quality inspections. This is particularly advantageous for complex building structures and large-scale projects.

[0032] Furthermore, the present application S100 includes:

[0033] Construct a geographic three-dimensional model of the building to be inspected and identify the building structure; decompose the building structure according to the quality inspection requirements to determine the inspection space structure, which has the quality inspection requirement identification characteristics; inspect the spatial characteristics and spatial connection relationship of the inspection space structure to obtain the spatial distribution relationship of quality inspection.

[0034] Furthermore, the present application further comprises the following steps:

[0035] The inspection space characteristics include the exterior and interior of the building. The spatial connection relationship represents the inspection space structural relationship and spatial connection position range between the exterior and interior of the building. The spatial distribution relationship of the quality inspection includes quality inspection requirements, inspection space characteristics and spatial relationships.

[0036] Specifically, data is collected using drone-mounted laser scanners or high-definition cameras. Combined with real estate engineering design drawings, detailed geometric data of the buildings to be inspected is collected, resulting in three-dimensional spatial data of the buildings. For example, architectural design drawings are obtained and LiDAR technology is used, combined with photogrammetry, using architectural design software such as Autodesk Revit or SketchUp to generate three-dimensional point cloud data of the building. This data is then converted into a visual three-dimensional model, accurately depicting spatial information such as the building's shape, structure, and layers. A geographic three-dimensional model is a three-dimensional digital representation of a building or geographic environment constructed using a variety of technologies (such as remote sensing, LiDAR, and drone photogrammetry). It includes not only information about the building's shape, dimensions, structure, and spatial layout, but also factors such as its geographic location and surrounding environment. The constructed geographic three-dimensional model identifies the building structure, including the elements that support and protect the building, such as walls, floors, roofs, beams, and columns.

[0037] Based on the building design, construction standards, and actual usage, different quality inspection requirements are defined. These are the items and inspection indicators that need to be inspected during the building quality inspection process, such as crack detection, leakage detection, and material defect detection. For each type of quality issue (such as cracks, leakage, and structural stability), the building structure is matched and decomposed according to different parts of the building structure, breaking it down into different inspection areas and marking the quality inspection requirements for each area, such as facade crack inspection, basement water seepage issues, and floor slab load-bearing capacity. For example, assuming the inspection task includes inspections of facade cracks, floor slab structural strength, and underground pipe leaks, the facade and basement areas of the building have different inspection requirements. The facade focuses on crack detection, while the basement focuses on waterproofing and pipe integrity inspections. The facade and basement areas are marked in the geographic 3D model.

[0038] The building space (i.e., the inspection space structure) is further subdivided to identify the inspection space characteristics and spatial connections of different spatial areas. Inspection space characteristics refer to the characteristics and attributes of different parts of the building space during the inspection process. For example, the exterior space of a building (facade, roof) and the interior space (corridors, offices, stairwells, etc.) have different inspection requirements and inspection methods. External inspections primarily focus on issues such as waterproofing, crack prevention, and wall structural stability. Internal inspections typically include structural safety, pipe installation, and wall cracks.

[0039] Spatial connectivity describes the interconnections and positional relationships between different building spaces, particularly the physical or functional connections between a building's exterior and interior. Spatial connectivity not only refers to physical connections (such as doors, windows, and passageways), but also includes functional connections between the exterior and interior (such as air conditioning systems and plumbing layouts). For example, connections between a building's exterior and interior, such as doors, windows, passageways, and air conditioning systems, possess certain physical or functional relationships. These connections must be identified during inspections to ensure comprehensiveness and accuracy.

[0040] By representing inspection space characteristics and spatial connectivity, different quality inspection requirements, inspection space characteristics, and spatial connectivity are combined to form a spatial distribution relationship for quality inspection. This helps to formulate inspection routes and adopt different inspection methods for different areas to ensure that quality issues in each space are fully inspected. The spatial distribution relationship of quality inspection refers to the determination of quality inspection requirements for different building areas based on inspection space characteristics and spatial connectivity, including quality inspection requirements, inspection space characteristics, and spatial relationships.

[0041] By constructing an accurate 3D building model and dividing inspection spaces into different areas based on quality inspection requirements, we ensure that every part of the building is fully inspected to avoid omissions. We also identify the connections between the building's interior and exterior spaces, enabling a more accurate understanding of the interactions between different areas and helping to identify potential structural issues. At the same time, we clarify the characteristics of inspection spaces and their connections, optimize inspection routes, avoid redundant inspections, and improve inspection efficiency.

[0042] Furthermore, the present application further comprises the following steps:

[0043] According to the multi-source acquisition characteristics of the sensors carried by the drone and combined with historical record data, the detection space characteristics and acquisition data characteristics of the multi-source sensors are determined, where the multi-source sensors include cameras, infrared thermal imagers, lidars, and ultrasonic sensors. According to the detection space characteristics and acquisition data characteristics of the multi-source sensors, the detection requirements are matched with the spatial distribution relationship of the quality detection, and an acquisition mapping of the multi-source sensors and the spatial distribution relationship is established. According to the acquisition mapping and the acquisition space constraints of the multi-source sensors, path planning is performed with the goal of full coverage of the detection spatial distribution relationship and path minimization to obtain the data source acquisition path.

[0044] Specifically, drones are equipped with multiple sensors, including cameras, infrared thermal imagers, lidar, and ultrasonic sensors, to collect different types of data to meet diverse inspection needs. For example, cameras capture high-definition images of building surfaces, infrared thermal imagers measure temperature variations, including thermal anomalies and insulation defects, lidar measures distances and generates 3D point cloud data, and ultrasonic sensors detect structural defects such as cavities and cracks.

[0045] Each sensor has different advantages and applicable scopes in different spatial characteristics. Understand the detection space characteristics of each sensor and the data characteristics they collect. Use the sensor's technical documentation and actual test data to analyze the performance of different sensors in different spatial characteristics, as well as the type and accuracy of the data they collect. At the same time, obtain the building's historical collection record data to understand the common problems that may occur in the building and their distribution characteristics. Based on the collection characteristics and historical records of multi-source sensors, determine the application scenarios of the sensors, including the sensor's detection space characteristics and collection data characteristics. Detection space characteristics refer to the characteristics of the spatial area or specific area that each sensor needs to pay attention to during the actual inspection process. Collection data characteristics refer to the type and attributes of the data that the sensor needs to collect.

[0046] Based on the detection space characteristics and data collection characteristics of multi-source sensors, a reasonable collection mapping is designed to match each sensor with specific areas and detection requirements in the building, ensuring that different sensors can effectively collect data at the corresponding spatial locations. In layman's terms, this means determining the location of each sensor to perform the collection task, determining the detection path and task allocation of the sensor in a specific area, and helping to rationally arrange the inspection tasks of different sensors, avoiding conflicts between sensors, and preventing omissions and duplicate collections. At the same time, it ensures that the quality inspection requirements of each area can be met and that all collected data is correct.

[0047] Each sensor has performance limitations, such as effective acquisition range, angle, and resolution. Furthermore, factors such as flight altitude, surrounding obstacles, and weather conditions can impact sensor acquisition. Based on these limitations, acquisition spatial constraints, including sensor performance constraints, environmental constraints, and safety constraints, are determined, resulting in the acquisition spatial constraints for the multi-source sensors. Based on the acquisition mapping and acquisition spatial constraints, full coverage of the detection spatial distribution relationship is ensured, while minimizing the path to reduce flight time and improve efficiency. In other words, with full quality inspection coverage and minimizing the drone's path as the goals, the acquisition mapping is used to determine the data source acquisition path, i.e., the drone's inspection path, within the constraints of the acquisition spatial constraints. The previously mentioned path planning algorithm is still used to optimize the path, taking into account all constraints, ensuring that each detection area is covered by at least one sensor, and considering the complementarity between sensors. Based on this, the drone's takeoff, landing, and emergency stops are planned.

[0048] Specifically, the A-star algorithm, based on the constraints of the acquisition space, selects the optimal path by calculating the shortest path from the starting point to each node and combining it with a heuristic evaluation function. Initialization begins by defining an open list and a closed list. The open list stores nodes to be explored, while the closed list stores nodes already explored. Each node is assigned an f-value, where f(n) = g(n) + h(n), where g(n) represents the actual cost (i.e., the path length) from the starting point to the current node, and h(n) represents the estimated cost (heuristic cost) from the current node to the target node, typically estimated using Euclidean distance or Manhattan distance. The node with the lowest f-value from the exploration list is selected as the current node and moved from the open list to the closed list. The f-values of all neighboring nodes of the current node are calculated and added to the open list. Each node's parent node points to its predecessor node, facilitating subsequent path backtracking. Once the target node is found, the optimal path from the starting point to the target node is obtained by tracing back each node's parent node.

[0049] The results of path planning should be translated into specific flight missions, and the drone's inspection mission should be executed. The drone will fly along the planned path, perform the corresponding inspection tasks, and collect data in real time. During flight, the drone's progress is monitored in real time to ensure that the drone is executing according to the set path and mission requirements. Through reasonable path planning and sensor allocation, the inspection tasks in each area can be completed efficiently, while redundant flights are avoided, ensuring that all quality inspection areas are covered without omission. This not only improves the efficiency and accuracy of drone quality inspections, but also optimizes the path planning and data collection processes.

[0050] Furthermore, the present application further comprises the following steps:

[0051] According to the spatial distribution relationship of the quality inspection, regional division is performed to obtain the drone collaborative collection area, wherein the regional division includes the division of the same plane space and the division of the opposite surfaces of the building space; the spatial detection constraint conditions between each area in the regional division result are obtained; according to the spatial detection constraint conditions, the drone collection path of each area is collaboratively planned to obtain the data source collection path, which includes multiple collaborative drone data source collection paths.

[0052] Specifically, when the building's spatial scope is large and the building environment is complex, using one drone to collect all the data is not only time-consuming and inefficient, but also the data transmission volume is too large, affecting the efficiency of drone inspections. In order to improve the efficiency of drone inspections and the comprehensiveness of data collection, a multi-drone collaborative collection strategy can be adopted to divide the real estate project into several areas, such as the north and south sides of the building. For example, if the area of the same floor is too large, it can be divided into multiple areas.

[0053] Based on the spatial distribution of quality inspections, the building's various areas, including exterior and interior areas, as well as individual floors and rooftops, are analyzed for quality inspection requirements and divided into different zones. For a single floor or roof, multiple zones are divided based on the plane. Data collection can be performed in parallel across these zones, reducing the burden on individual drones. For building facades, they can be divided into different facets, such as the east and west facades. Each facet can be assigned a drone to ensure thorough inspection. For example, suppose a 30-story building requires crack inspection on each floor, camera inspection on the facade, and temperature measurement on the roof using an infrared thermal imaging camera. Each floor is divided into a monitoring zone, which is then further divided into four zones: the east, west, south, and north facades, with four zones each. The rooftop area is a separate inspection zone.

[0054] Area division includes same-plane space division and building space opposite-surface division. Same-plane space division refers to dividing the building according to a certain plane (such as ground plane, floor plane, roof plane, etc.), which is suitable for horizontal areas of the building, such as inspection tasks between floors; building space opposite-surface division is based on different spatial surfaces such as the building's facade, wall, roof, etc., and is suitable for structural inspections of the building's facade, roof, etc.

[0055] Each detection area has different spatial constraints, which affect the drone's path planning. Different sensors have different flight altitude requirements. For example, cameras require the drone to be stable, while ultrasonic sensors require a lower altitude. Based on the drone's safe flight distance, obstacle avoidance requirements, sensor coverage, data overlap areas, and the characteristics of the divided areas, the constraints that the drone must adhere to when collecting data are determined, namely the spatial detection constraints.

[0056] Based on spatial constraints, drone collection paths are planned for each area. Each drone collects data along the assigned path to ensure coverage of all key inspection points. Path planning here is similar to the previous section and will not be detailed for the sake of brevity. The difference lies in determining inspection priorities based on the defects to be inspected and the timeliness of inspections. The weights for the corresponding inspection targets are then configured accordingly, allowing the drone collection paths to be modified. When multiple drones are working together, each drone's path should not interfere with other drones, and the sensors in each area should be able to perform their tasks at the appropriate location and angle.

[0057] For example, assume there are four drones, each covering four areas. Each area has a 50-meter flight path, and the inspection data includes image data, temperature data, and crack monitoring data. Drone 1 is used in Area 1, with a flight path of 60 meters, a flight time of 6 seconds, 60 MB of collected data, and a data transmission time of 6 seconds, resulting in an inspection time of 12 minutes. Drone 2 is used in Area 2, with a flight path of 80 meters, a flight time of 8 seconds, 80 MB of collected data, and a data transmission time of 8 seconds, resulting in an inspection time of 15 minutes. Drone 3 is used in Area 3, with a flight path of 70 meters, a flight time of 7 seconds, 80 MB of collected data, and a data transmission time of 8 seconds, resulting in an inspection time of 14 minutes. Drone 4 is used in Area 4, with a flight path of 100 meters, a flight time of 10 seconds, 110 MB of collected data, and a data transmission time of 10 seconds, resulting in an inspection time of 18 minutes. Therefore, when multiple drones perform an inspection, the total time taken is no more than 20 minutes, which is significantly more efficient than a single drone, which takes nearly an hour. At the same time, multiple drones work together according to the planned route, making the detection task in each area independent and efficient, and avoiding interference between drones.

[0058] When performing inspection tasks, drones fly according to the revised collection paths to ensure effective coverage of all areas, taking into account the coordinated operation of multiple drones. The data source collection path includes multiple coordinated drone data source collection paths—that is, the flight paths and data collection tasks of multiple drones working together. Through collaborative planning, each drone's path does not interfere with each other, ensuring that inspection tasks in each area are properly allocated and all inspection points are fully covered. The coordinated operation of multiple drones can simultaneously cover multiple areas of a building, reducing the flight time of a single drone and improving data collection efficiency.

[0059] Furthermore, the present application further comprises the following steps:

[0060] According to the quality inspection requirements corresponding to the spatial inspection constraints, a quality inspection priority is obtained, where the quality inspection priority is proportional to the impact of quality defects and the timeliness of inspection. According to the quality inspection priority, the collaborative planning weights of the spatial inspection constraints are configured. Based on the spatial inspection constraints and their collaborative planning weights, the drone collection paths in each area are collaboratively planned.

[0061] Specifically, each quality inspection task has a corresponding impact and timeliness. Tasks with greater impact and a shorter timeliness are generally given higher priority. Quality inspection priorities are determined by assessing the impact of different quality defects on the building's function, safety, and service life. Structural cracks are generally given a higher priority than surface coating defects. For example, structural cracks (which impact building safety and have a high timeliness) and water pipe leaks (which can cause serious damage and require prompt inspection) are given a high priority, while surface cracks (which affect appearance and can be inspected later) and temperature change inspections (general inspections with a lower timeliness) are given a low priority.

[0062] Quality inspection priority is typically determined by the impact of a quality defect and its timeliness. Tasks with greater impact or timeliness are given higher priority. The impact of a quality defect refers to the potential impact of a defect on a building's safety, performance, and appearance. Timeliness refers to the time sensitivity of a quality inspection task. Tasks with high timeliness may be prioritized. For example, crack detection, which poses a safety hazard, requires expedited completion, while temperature monitoring may be more flexible.

[0063] Based on the impact of the defect and the timeliness of detection, a weighted formula or scoring system is used to calculate the priority of each detection task. i =w1·I i +w2·T i , where P iis the priority of the i-th task, w1 is the weight coefficient of the impact of quality defects, w2 is the weight coefficient of detection timeliness, I i and T i are the impact score and timeliness score of the i-th task, respectively. For example, assuming the impact score of the building crack inspection task is 8 and the timeliness score is 7, and the impact score of the building surface crack inspection task is 5 and the timeliness score is 2, and the quality defect impact weight and inspection timeliness weight are set to 0.7 and 0.3 respectively, then the priority of the crack inspection task is 0.7×8+0.3×7=7.7, and the priority of the surface crack inspection task is 0.7×5+0.3×2=4.1, indicating that the crack inspection task has a higher priority.

[0064] Based on the quality inspection priority, the task priority of each area is combined with the spatial inspection constraints to generate corresponding collaborative planning weights. This not only reflects the importance (priority) of the task, but also incorporates the task's spatial inspection constraints (such as flight altitude, sensor operating range, etc.). For example, when there are conflicting spatial constraints, such as data from multiple locations that need to be collected simultaneously on both sides of a building, but there are spatial position deviations and they cannot be satisfied simultaneously, the weights of the corresponding inspection targets are considered. Path planning is performed first for those that require immediate inspection and have high requirements, while less urgent data can be processed through subsequent spatial coordinate alignment.

[0065] When data needs to be collected simultaneously on both sides of a building, conflicting spatial constraints may arise. For example, drones on either side may need to collect data at the same location at the same time, but their flight paths may differ. In complex environments, such as those surrounding buildings with obstacles or no-fly zones, drone flight paths may be restricted, leading to conflicting spatial constraints. Based on all these conditions, collaborative planning weights are configured, assigning each inspection task in each area a value that reflects the importance and urgency of the inspection task in that area.

[0066] Drone paths are planned based on the collaborative planning weight for each area. High-priority and high-weight areas are prioritized to ensure the most urgent and important tasks are completed first. Furthermore, overlapping flight paths of multiple drones are avoided, ensuring that tasks in each area are not duplicated and that coverage is efficient. Specifically, paths are prioritized for high-weight areas based on the collaborative planning weights of the areas. Each drone is assigned a specific mission area, and its path is planned based on the spatial constraints of that area. Tasks are intelligently assigned to different drones based on their priority and weight. Given the complexity of the building, path planning aims to avoid collisions and intersections between drones to ensure flight safety and efficiency. The aforementioned path planning details apply equally here and are not detailed here. Collaborative planning enables the simultaneous use of multiple drones for data collection, significantly improving collection efficiency and reducing flight time and costs. At the same time, it ensures that all areas are covered, enhancing the comprehensiveness and accuracy of data collection.

[0067] S200: Positioning the collected data according to the data source collection path, and constructing data space coordinates.

[0068] Furthermore, the present application S200 includes:

[0069] The structural position of the building to be inspected where the inspection data is located is located according to the data source acquisition path to obtain the building positioning coordinates of the inspection data; based on the spatial relationship of the building structure, a spatial coordinate system is constructed; the building positioning coordinates of all inspection data are projected into the spatial coordinate system to obtain the data spatial coordinates, wherein the inspection data has sensor type and quality inspection requirements.

[0070] Specifically, the data source collection path includes path planning for each drone across multiple areas. Different drones have different flight paths in different areas. Drones collect data according to the data source collection path, generating inspection data. Inspection data refers to the raw data collected by drones using their multi-source sensors during inspections. This data includes information about a building's appearance, structure, temperature distribution, cracks, and other aspects, and is determined by sensor type and quality inspection requirements.

[0071] By analyzing the inspection data, the specific location of each data point within the building being inspected is determined. Key building structures and features, such as walls, floors, and columns, are identified and correlated with the drone's position data to obtain the building coordinates of the inspection data. In other words, the collected inspection data is correlated with specific locations in the building, and the specific spatial coordinates of the data point on the building are obtained through the drone's flight path and position.

[0072] Based on the spatial relationship of the building structure, including the relative position and geometry between the building's external space (facade, floors, roof) and the building's internal space (room layout, wall thickness, windows, etc.), the building's spatial layout, functional zoning, connection relationship and the relative position of each structural part in three-dimensional space are defined.

[0073] In a building's spatial coordinate system, a reference point (origin) is first selected. This is usually a specific location within the building, such as a corner, the center of a room, or the building's foundation. Based on the building's layout, the directions of the X, Y, and Z axes are defined. Each floor can be represented by a different Z value. Inside a building, there are various functional areas, such as lobbies, offices, and corridors. Each area can occupy a specific spatial range in the spatial coordinate system. The spatial coordinate system is determined by the distance from the reference point, describing the relative positional relationships between areas. The building's external structures, such as the facade and roof, also need to be calibrated in the spatial coordinate system. For example, the coordinates of the roof may correspond to the building's highest Z value, while the facade walls extend along the building's X and Y directions.

[0074] Furthermore, each architectural element, such as a room, wall, window, and door, occupies a specific position in the spatial coordinate system and has clear spatial relationships. The spatial coordinate system describes the relative positions of these elements. In complex buildings, connections may exist between various spatial units, such as doors and corridors. The spatial coordinate system allows these connections to be described, allowing the physical connections between different areas to be reflected in the coordinate system. In general, a building's physical structure, internal layout, and external environment are fully displayed in the spatial coordinate system.

[0075] Map each collected data point (such as image, temperature, crack depth, etc.) to the building positioning coordinates of its location to form a data positioning relationship. Project each data point to the corresponding location based on the building's spatial coordinate system. The data collected by each sensor has different characteristics. How to organize this data in the spatial coordinate system depends on the data type and quality inspection requirements of different sensors. Once all collected inspection data are projected into the spatial coordinate system, a complete data space coordinate can be formed, including the spatial location of each collected data point and the corresponding quality inspection requirements (such as crack detection, temperature monitoring, etc.). By combining inspection data with building positioning coordinates and the spatial coordinate system, the accuracy of data positioning can be ensured, and clear spatial visualization can be provided to quickly locate defects in the building and improve inspection efficiency.

[0076] S300: Establishing a quality detection category framework, wherein the quality detection category framework includes quality detection categories, data requirement parameters, identification detection operators, and detection paths.

[0077] Furthermore, the present application S300 includes:

[0078] According to the detection timeliness of each quality detection requirement, a detection timeliness hierarchy is established; according to the data requirement parameters of each quality detection requirement and the processing process of the identification detection operator, the processing relationship is fitted with the multi-source collected data of the drone to obtain the processing identification path of each quality detection requirement; based on the processing identification path of each quality detection requirement, the data transmission requirement and computing power requirement of each path node are obtained, and an operation requirement chain is established; with the detection time threshold of each quality detection requirement in the detection timeliness hierarchy as a constraint, the edge device node of the operation requirement chain is matched to construct the quality detection category framework, which is used to reflect the edge device node path and data requirement parameters and identification detection operator corresponding to the quality detection requirement during the processing process.

[0079] Specifically, analyze all quality inspection requirements, including crack detection, temperature detection, structural integrity inspection, etc. Each quality inspection requirement has different inspection objectives, data requirements, and inspection methods. Based on each quality inspection requirement, assess the urgency and importance of each requirement and assign a timeliness level to each quality inspection requirement. High-timeliness tasks need to be completed first, while low-timeliness tasks can be appropriately deferred. For example, structural cracks require rapid identification of possible structural damage and have a high timeliness level; temperature detection is more important but does not immediately threaten the safety of the building and has a medium timeliness level; facade cleaning inspections are more routine and can be delayed, so they have a low timeliness level.

[0080] For each quality inspection requirement, determine the required data requirement parameters and identification detection operators. Data requirement parameters represent the specific data types (such as temperature, image, and depth information) and data collection methods required for each quality inspection requirement. Identification detection operators are algorithms or tools used to analyze and process collected data. Different quality inspection requirements may require different operators for data analysis. For example, crack detection may require an image processing algorithm, while temperature monitoring may require a thermal imaging analysis algorithm. Furthermore, analyze the data types and features provided by the multi-source sensors onboard drones (such as cameras and infrared thermal imagers).

[0081] By fitting the processing relationship between each quality inspection requirement and the corresponding multi-source sensor data, a processing and identification path for each quality inspection requirement is obtained. This process shows the entire process from data acquisition to quality inspection result generation, including steps such as data preprocessing, feature extraction, and detection and identification. The fitting process involves determining the mutual matching of acquired data, recognition operators, and data processing to ensure that each quality inspection task is completed in an optimal manner. Each processing and identification path includes data requirements (clarifying the specific data type required for each quality inspection requirement), recognition operators (determining the algorithm or operator selected for each inspection task), data transmission and processing paths (the process from data acquisition to transmission and processing), and sensor matching (ensuring that the sensor can provide the required data). For example, the processing and identification path for crack detection includes data acquisition (image data), data transmission, edge device processing (edge detection algorithm), and detection result output (crack location, size, etc.).

[0082] Based on the processing identification path for each quality inspection requirement, the computational requirements of each node in the path are further determined, including data transmission rate and bandwidth. Each task requires different data transmission and computing resources, so a computational requirements chain is needed to coordinate the requirements of each node. For each quality inspection requirement, the transmission requirements (such as bandwidth and transmission speed) and computational requirements (such as CPU processing power and storage requirements) after data acquisition are calculated. These requirements are combined to form a computational chain (computational requirements chain). Each node represents the data acquisition, data transmission, and data processing stages. Data transmission requirements describe how the drone transmits collected data to back-end systems or devices during the inspection process, including parameters such as data transmission speed and bandwidth. Computational processing power requirements are required to process the data generated during the inspection process, including the computing power and storage requirements of the computing nodes. All nodes in the processing identification path are connected according to the data processing flow to form a computational requirements chain, which reflects the data flow from data acquisition to quality inspection result generation and the computational resources required at each node.

[0083] The detection time threshold of each quality detection requirement in the detection time efficiency level is used as a constraint. That is to say, according to the time threshold defined in the detection time efficiency level, the processing time limit is determined for each quality detection requirement. Match each node of the computing demand chain with a suitable edge device, and assign appropriate tasks to each device node. During the matching process, it is necessary to ensure that each task can be completed within the specified time to avoid missing the time efficiency requirements due to insufficient resources or processing delays. For high-time efficiency tasks, it is necessary to select edge device nodes with strong processing capabilities and fast response times; for low-time efficiency tasks, edge device nodes with lower computing power can be selected for processing.

[0084] By organizing and integrating various elements of quality inspection tasks, including timeliness requirements, data requirements, and recognition operators, a complete quality inspection category framework is constructed. This framework reflects the edge device paths, data requirement parameters, and recognition operators required for each task. After establishing this quality inspection category framework, an intelligent scheduling system optimizes multiple tasks. Based on timeliness priorities, edge device node capabilities, and computing resources, the task processing sequence and resource allocation are rationally arranged to ensure that each task is completed efficiently within the specified timeframe.

[0085] The quality inspection category framework defines the processing flow, data requirement parameters, identification and detection operators, and required edge device resources for each quality inspection requirement. This provides detailed operational guidance for implementing drone inspections, ensuring the accuracy and timeliness of quality inspections. By dividing quality inspection requirements into time-sensitive categories and scheduling tasks, high-priority tasks can be completed promptly and unnecessary resource waste can be avoided. The quality inspection category framework provides a structured management approach that can address complex and diverse quality inspection requirements, improve overall task processing efficiency, and thus significantly enhance the efficiency and accuracy of building quality inspections.

[0086] S400: Based on the data space coordinates, the drone inspection data is input into the quality inspection category framework for inspection path matching and import, the inspection data is identified and extracted through the data requirement parameters, and the inspection category identification processing is performed using the identification detection operator to obtain the quality inspection result of the corresponding quality inspection category.

[0087] Specifically, the data collected by drone inspections is integrated based on data space coordinates and input into the quality inspection classification framework. The framework then matches the corresponding data requirement parameters, identifies detection operators, and selects the appropriate edge devices. Data extraction and processing are performed based on the preset data requirement parameters. The quality inspection classification framework extracts the required data features from the integrated inspection data. The extracted data is processed using the identification detection operators corresponding to each quality inspection task. For example, image recognition algorithms (such as edge detection and image segmentation) can be used to identify cracks in walls or floors.

[0088] After processing by the recognition and detection operators, the results of each quality inspection task are output, clearly indicating quality issues in the building, such as crack locations and areas of abnormal temperature. Depending on the quality inspection task, the results may be presented in different forms, such as image annotation, temperature distribution maps, and structural defect reports. By precisely associating inspection data with the spatial location of the building through data space coordinates, the accuracy of data processing and quality inspection results is ensured. By matching different types of inspection data with the requirements of quality inspection tasks, quality issues in the building, such as cracks and temperature anomalies, can be efficiently identified, providing more accurate inspection results, timely discovering quality issues in the building, reducing building hazards, and improving building safety.

[0089] Furthermore, the present application S400 includes:

[0090] Data alignment and integration are performed according to the data space coordinates, sensor type, and quality inspection requirements of the inspection data; the integrated inspection data is input into the quality inspection category framework, matched with the corresponding edge device nodes, and quality inspection of the corresponding building structure position is performed according to the data space coordinates.

[0091] Specifically, inspection data is aligned and integrated based on the spatial coordinates of the data obtained from the inspection data, the various sensor types involved in the inspection data, and the quality inspection requirements. Inspection data from different sensors and different locations is uniformly integrated according to the building's spatial coordinates and inspection requirements, so that all data can be processed in the same reference system. In other words, during the integration process, different types of data need to be matched with the corresponding building locations to ensure that they can be processed synchronously in the same coordinate system. According to the predetermined quality inspection tasks (such as crack detection, temperature anomaly detection, structural analysis, etc.), the corresponding sensor data and quality inspection requirements are matched. Each task has different data requirement parameters and detection algorithms, so the data needs to be screened and prepared according to the requirements.

[0092] Once all sensor data is aligned with quality inspection requirements, data alignment and integration are required. Using a data space coordinate system, all data (including data collected by different sensors) is uniformly processed to ensure that all data accurately corresponds to the building's spatial location. Aligning the data from each sensor with the building's 3D model ensures that data from different sensors can be processed in the same spatial coordinate system, avoiding data mismatches caused by inconsistent coordinates.

[0093] The integrated data is fed into the quality inspection category framework. Based on pre-defined quality inspection requirements (such as crack detection and temperature monitoring), the framework matches the required sensor data and recognition operators. Based on the timeliness and computational requirements of the task, the framework selects the appropriate edge device node for processing. The framework also ensures that each piece of data processed corresponds to the specific location of the building, using a spatial coordinate system to perform quality inspections on the building structure.

[0094] Based on the matching results in the quality inspection category framework, the edge device will begin executing the quality inspection task, processing, analyzing, and providing feedback on the data for each area and location. For example, for the crack detection task, the edge device will use image data to execute a crack recognition algorithm to identify cracks on the building's exterior wall and output the crack's location and severity. For the temperature detection task, the edge device will analyze infrared data to detect abnormal temperature areas on the building surface and indicate whether there is heat loss or overheating.

[0095] By aligning the data from different sensors with the building's spatial coordinate system, ensuring that all data can be accurately mapped to the specific location of the building structure, avoiding data mismatch and confusion, the quality inspection category framework can automatically select appropriate data processing methods and edge device nodes according to task requirements, making the quality inspection process more efficient and accurate, and flexibly responding to different types of quality inspection tasks. It can also be adjusted in real time according to the actual situation of the building, significantly improving the data processing efficiency and accuracy during the quality inspection process.

[0096] In summary, the real estate project quality inspection method based on drone inspection provided by this application has the following beneficial effects:

[0097] The invention relates to a method for accurately locating the data source collection path based on the spatial distribution of quality inspections. The method locates the collected data according to the data source collection path and constructs data spatial coordinates. The method establishes a quality inspection category framework, which includes quality inspection categories, data requirement parameters, identification detection operators, and inspection paths. The method then inputs the drone inspection data into the quality inspection category framework based on the data spatial coordinates for inspection path matching and importing. The inspection data is identified and extracted using the data requirement parameters, and the identification detection operators are used to perform inspection category identification processing to obtain quality inspection results corresponding to the quality inspection categories. In other words, by analyzing the spatial distribution of quality inspections and performing path planning, the drone-collected data is accurately located with the building structure, reducing errors caused by interference factors and ensuring that each piece of data truly reflects the actual structural state. The method constructs a comprehensive and efficient quality inspection category framework. After the inspection data is input into the quality inspection framework, the detection operators are used to automatically identify and output quality inspection results. This overcomes the limitation of drones in detecting small or complex defects in the quality inspection of high-rise buildings and complex components, thereby improving the accuracy and efficiency of real estate project quality inspections.

[0098] In the second embodiment, based on the same inventive concept as the real estate project quality inspection method based on drone inspection in the first embodiment, this application also provides a real estate project quality inspection system based on drone inspection, please refer to the attached Figure 2 The real estate project quality inspection system based on drone inspection includes:

[0099] The path planning module 11 is used to plan the UAV inspection path according to the spatial distribution relationship of quality inspection and obtain the data source collection path; the coordinate positioning module 12 is used to locate the collected data according to the data source collection path and construct the data space coordinates; the framework construction module 13 is used to establish a quality inspection category framework, and the quality inspection category framework includes quality inspection category, data requirement parameters, identification detection operator and inspection path; the inspection result generation module 14 is used to input the UAV inspection data into the quality inspection category framework based on the data space coordinates to perform inspection path matching and import, identify and extract the inspection data through the data requirement parameters, and use the identification detection operator to perform inspection category identification processing to obtain quality inspection results corresponding to the quality inspection category.

[0100] Furthermore, the path planning module 11 in the real estate project quality inspection system based on drone inspection is also used to: construct a geographical three-dimensional model of the building to be inspected and identify the building structure; decompose the building structure according to the quality inspection requirements to match the requirements and determine the inspection space structure, and the inspection space structure has quality inspection requirement identification characteristics; identify the inspection space characteristics and spatial connection relationships of the inspection space structure to obtain the spatial distribution relationship of quality inspection.

[0101] Furthermore, the path planning module 11 in the real estate project quality inspection system based on drone inspection is also used for: the inspection space characteristics include the exterior and interior of the building, the spatial connection relationship represents the inspection space structural relationship and spatial connection position range between the exterior and interior of the building, and the spatial distribution relationship of the quality inspection includes quality inspection requirements, inspection space characteristics and spatial relationships.

[0102] Furthermore, the path planning module 11 in the real estate project quality inspection system based on drone inspection is also used to: determine the detection space characteristics and acquisition data characteristics of the multi-source sensors according to the multi-source acquisition characteristics of the sensors carried by the drone, combined with historical record data, wherein the multi-source sensors include cameras, infrared thermal imagers, lidars, and ultrasonic sensors; according to the detection space characteristics and acquisition data characteristics of the multi-source sensors, match the detection requirements with the spatial distribution relationship of the quality inspection, and establish an acquisition mapping of the multi-source sensors and the spatial distribution relationship; according to the acquisition mapping combined with the acquisition space constraints of the multi-source sensors, path planning is performed with the goal of full coverage of the detection spatial distribution relationship and path minimization to obtain the data source acquisition path.

[0103] Furthermore, the path planning module 11 in the real estate project quality inspection system based on drone inspection is also used to: divide the area according to the spatial distribution relationship of the quality inspection to obtain the drone collaborative collection area, wherein the area division includes the same plane space division and the opposite surface division of the building space; obtain the spatial detection constraint conditions between each area in the area division result; according to the spatial detection constraint conditions, collaboratively plan the drone collection path of each area to obtain the data source collection path, and the data source collection path includes multiple collaborative drone data source collection paths.

[0104] Furthermore, the path planning module 11 in the real estate project quality inspection system based on drone inspection is also used to: obtain the quality inspection priority according to the quality inspection requirements corresponding to the spatial inspection constraints, and the quality inspection priority is proportional to the impact of quality defects and the timeliness of inspection; configure the collaborative planning weights of the spatial inspection constraints according to the quality inspection priority; and collaboratively plan the drone collection paths of each area based on the spatial detection constraints and their collaborative planning weights.

[0105] Furthermore, the coordinate positioning module 12 in the real estate project quality inspection system based on drone inspection is also used to: locate the structural position of the building to be inspected where the inspection data is located according to the data source acquisition path, and obtain the building positioning coordinates of the inspection data; construct a spatial coordinate system based on the spatial relationship of the building structure; project the building positioning coordinates of all inspection data into the spatial coordinate system to obtain the data spatial coordinates, wherein the inspection data has sensor type and quality inspection requirements.

[0106] Furthermore, the framework construction module 13 in the real estate project quality inspection system based on drone inspection is also used to: establish a detection timeliness hierarchy according to the detection timeliness of each quality inspection requirement; perform processing relationship fitting with the multi-source collected data of the drone according to the data requirement parameters and the processing process of the identification detection operator of each quality inspection requirement, and obtain the processing identification path of each quality inspection requirement; obtain the data transmission requirements and computing power requirements of each path node based on the processing identification path of each quality inspection requirement, and establish an operation requirement chain; use the detection time threshold of each quality inspection requirement in the detection timeliness hierarchy as a constraint to match the edge device nodes of the operation requirement chain and construct the quality inspection category framework, which is used to reflect the edge device node path and data requirement parameters and identification detection operators corresponding to the quality inspection requirements during the processing process.

[0107] Furthermore, the inspection result generation module 14 in the real estate project quality inspection system based on drone inspection is also used to: align and integrate data according to the data space coordinates, sensor type, and quality inspection requirements of the inspection data; input the integrated inspection data into the quality inspection category framework, match the corresponding edge device nodes, and perform quality inspection of the corresponding building structure position according to the data space coordinates.

[0108] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. Figure 1The real estate project quality inspection method based on drone inspection and the specific examples in Example 1 are also applicable to the real estate project quality inspection system based on drone inspection in this embodiment. Through the above detailed description of the real estate project quality inspection method based on drone inspection, those skilled in the art can clearly understand the real estate project quality inspection system based on drone inspection in this embodiment, so for the sake of brevity of the specification, it will not be described in detail here.

[0109] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present application. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application is not limited to the embodiments shown herein, but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

[0110] Obviously, for those skilled in the art, several improvements and modifications can be made to the present application without departing from the principles of the present application, and these improvements and modifications also fall within the scope of protection of the present application.

Claims

1. A real estate project quality inspection method based on drone inspection is characterized by: include: According to the spatial distribution relationship of quality inspection, the drone inspection path is planned to obtain the data source collection path; Locate the collected data according to the data source collection path and construct data space coordinates; Establishing a quality detection category framework, wherein the quality detection category framework includes quality detection categories, data requirement parameters, identification detection operators, and detection paths; Based on the data space coordinates, the drone inspection data is input into the quality inspection category framework for inspection path matching and import, the inspection data is identified and extracted using the data requirement parameters, and the inspection category is identified and processed using the identification detection operator to obtain the quality inspection result corresponding to the quality inspection category; The establishment of a quality inspection category framework includes: Establish a testing timeliness hierarchy based on the testing timeliness of each quality testing requirement; According to the data requirement parameters of each quality inspection requirement and the processing process of the identification detection operator, the processing relationship is fitted with the multi-source collected data of the UAV to obtain the processing and identification path of each quality inspection requirement; Based on the processing identification path of each quality detection requirement, the data transmission requirement and computing power requirement of each path node are obtained, and a computing requirement chain is established; Taking the detection time threshold of each quality detection requirement in the detection time efficiency level as a constraint, the edge device node matching is performed on the operation requirement chain to construct the quality detection category framework. The quality detection category framework is used to reflect the edge device node path and data requirement parameters corresponding to the quality detection requirement during the processing process and identify the detection operator.

2. The real estate project quality detection method based on drone inspection according to claim 1 is characterized in that: The UAV inspection path planning according to the spatial distribution relationship of quality inspection previously includes: Construct a geographical 3D model of the building to be inspected and identify the building structure; Decomposing the building structure according to the quality inspection requirements to match the requirements, and determining an inspection space structure, wherein the inspection space structure has a quality inspection requirement identification feature; The inspection space structure is inspected for spatial features and spatial connection relationship identification and marking to obtain the spatial distribution relationship of quality inspection.

3. The real estate project quality detection method based on drone inspection according to claim 2 is characterized in that: The inspection space characteristics include the exterior and interior of the building. The spatial connection relationship represents the inspection space structural relationship and spatial connection position range between the exterior and interior of the building. The spatial distribution relationship of the quality inspection includes quality inspection requirements, inspection space characteristics and spatial relationships.

4. The real estate project quality detection method based on drone inspection according to claim 2 is characterized in that: The obtaining of the data source acquisition path includes: According to the multi-source collection characteristics of the sensors carried by the UAV and combined with historical data, the detection space characteristics and collection data characteristics of the multi-source sensors are determined. The multi-source sensors include cameras, infrared thermal imagers, lidars, and ultrasonic sensors. According to the detection space characteristics and collected data characteristics of the multi-source sensor, the detection requirements are matched with the spatial distribution relationship of the quality detection, and an acquisition mapping between the multi-source sensor and the spatial distribution relationship is established; According to the acquisition mapping and the acquisition space constraints of the multi-source sensor, path planning is performed with the goal of full coverage of the detection spatial distribution relationship and path minimization to obtain the data source acquisition path.

5. The real estate project quality detection method based on drone inspection according to claim 4 is characterized in that: The obtaining of the data source acquisition path further includes: According to the spatial distribution relationship of the quality inspection, regional division is performed to obtain the drone collaborative collection area, wherein the regional division includes the same plane space division and the opposite surface division of the building space; Obtaining spatial detection constraints between regions in the region division result; The UAV collection paths in each area are collaboratively planned according to the spatial detection constraint conditions to obtain the data source collection path, which includes multiple collaborative UAV data source collection paths.

6. The real estate project quality inspection method based on drone inspection according to claim 5 is characterized in that: Collaborative planning of drone collection paths in each area based on the spatial detection constraints also includes: Obtaining a quality inspection priority according to the quality inspection requirement corresponding to the spatial inspection constraint, where the quality inspection priority is proportional to the impact of quality defects and the timeliness of inspection; configuring collaborative planning weights of the spatial detection constraints according to the quality detection priorities; Based on the spatial detection constraints and their collaborative planning weights, collaborative planning is performed on the UAV collection paths in each area.

7. The real estate project quality inspection method based on drone inspection according to claim 1 is characterized in that: Based on the data space coordinates, the drone inspection data is input into the quality inspection category framework for inspection path matching and import, including: Align and integrate data according to the data space coordinates, sensor type, and quality inspection requirements of the inspection data; The integrated inspection data is input into the quality inspection category framework, matched with the corresponding edge device node, and quality inspection of the corresponding building structure position is performed according to the data space coordinates.

8. The real estate project quality inspection method based on drone inspection according to claim 4 is characterized in that: Construct data space coordinates, including: Locate the structural position of the building to be inspected where the inspection data is located according to the data source acquisition path, and obtain the building positioning coordinates of the inspection data; Construct a spatial coordinate system based on the spatial relationship of the building structure; The building positioning coordinates of all inspection data are projected into the spatial coordinate system to obtain the data spatial coordinates, wherein the inspection data has sensor type and quality detection requirements.

9. The real estate project quality inspection system based on drone inspection is characterized by: The method for implementing the real estate project quality inspection method based on drone inspection according to any one of claims 1 to 8 is provided, wherein the real estate project quality inspection system based on drone inspection comprises: The path planning module is used to plan the inspection path of the UAV according to the spatial distribution relationship of the quality inspection and obtain the data source collection path; A coordinate positioning module is used to locate the collected data according to the data source collection path and construct the data space coordinates; A framework construction module is used to establish a quality detection category framework, wherein the quality detection category framework includes quality detection categories, data requirement parameters, identification detection operators and detection paths; The detection result generation module is used to input the drone inspection data into the quality inspection category framework based on the data space coordinates to perform detection path matching and import, identify and extract the inspection data through the data requirement parameters, and use the recognition detection operator to perform detection category recognition processing to obtain the quality inspection results of the corresponding quality inspection category.

Images