Inspection method and system for engineering construction

Abstract

The invention relates to the technical field of unmanned aerial vehicle flight inspection control, and discloses an engineering construction inspection method and system, and the system comprises an engineering construction inspection field data collection module, an unmanned aerial vehicle inspection path planning module, an unmanned aerial vehicle inspection obstacle avoidance management module, and an engineering construction inspection quality problem point data collection module. Performing image feature matching processing on the construction inspection site feature image data and the construction inspection site quality problem image data by adopting a bee colony algorithm, and identifying quality problem points of the construction inspection site; and when identifying that the quality problem point exists in the construction inspection site, recording and collecting the identified quality problem image data of the construction inspection site and the inspection position space coordinates of the construction inspection site, thereby improving the identification efficiency and accuracy of the quality problem point in engineering construction inspection operation.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned aerial vehicle (UAV) flight inspection control, and in particular to a method and system for engineering construction inspection. Background Art

[0002] Unmanned aerial vehicles, also known as "drones" or "UAVs" in English, are unmanned aircraft controlled by radio remote control equipment and self-contained program control devices, or operated completely or intermittently autonomously by on-board computers. Drones can be divided into military and civilian use according to their fields of application. In the military, drones are divided into reconnaissance aircraft and target aircraft. In the civilian sector, drones + industry applications are the real rigid demand for drones; their applications in aerial photography, agriculture, plant protection, micro selfies, express delivery, disaster relief, wildlife observation, infectious disease monitoring, surveying and mapping, news reporting, power inspections, disaster relief, film and television shooting, etc., have greatly expanded the use of drones themselves; for the heavy workload and dangerous construction engineering inspection tasks, drones can complete the inspection tasks efficiently and with high quality compared to manual inspections. Construction inspections can ensure that personnel, materials, equipment, construction progress and quality are under control during the construction process, discover defects and potential safety hazards in construction operations in advance, and take effective measures in a timely manner to ensure the safety of construction. ; Existing drones for construction inspections mostly rely on manual inspections, and collecting construction inspection data requires manual identification of quality problem points at the construction site, which is labor-intensive and inefficient. With the development of technology, intelligent path planning algorithms are used to plan paths for drones, and construction inspection tasks are carried out in conjunction with static construction site three-dimensional models, enabling autonomous flight and data collection of drones during the construction inspection process. However, the construction inspection site not only has static buildings, but also constantly moving construction personnel and construction equipment. Drones for construction inspections need to accurately avoid obstacles, and at the same time autonomously identify quality problem points during construction inspections and record their locations.

[0003] Chinese application publication number CN114706419A discloses a BIM-based construction site drone collaborative inspection scheduling method and system. This method utilizes BIM-based time dimensions, the three-dimensional space of the construction process, and monitoring demand information, leveraging the advantages of drones to conduct comprehensive real-time monitoring and feedback information for managers to make timely decisions. This method can achieve the coordination of physical construction and computer virtual construction, guiding the effective, safe, and high-quality implementation of the construction process. Taking into account multiple factors such as inspection needs, spatial location, and monitoring quality, a two-layer algorithm for multi-drone collaborative inspection paths is proposed. This algorithm efficiently generates drone inspection task allocation schemes, inspection path planning schemes, and inspection time and speed control methods to optimize the quality of construction site inspections. This scheme can efficiently complete construction inspections at construction sites with large on-site spaces, numerous quality inspection tasks, and higher quality requirements. However, drones cannot accurately avoid obstacles for construction workers and equipment that are constantly moving on the construction site, nor can they autonomously identify quality issues during construction inspections or record their locations. Summary of the Invention

[0004] (1) Technical problems solved

[0005] In order to solve the problem that the above-mentioned construction inspection site not only has static buildings, but also constantly moving construction personnel and construction equipment, drones need to accurately avoid obstacles when conducting construction inspections, and at the same time autonomously identify quality problem points during construction inspections and record the locations of quality problem points. The above purpose is to achieve flight obstacle avoidance for dynamic construction personnel, construction equipment, and buildings based on safety distance threshold feedback, and to identify and record data and locations of quality problem points in construction projects based on image recognition.

[0006] (2) Technical solution

[0007] The present invention is implemented through the following technical solution: a method for engineering construction inspection, the method comprising the following steps:

[0008] S1. Scan and collect three-dimensional features of the construction inspection site to generate three-dimensional feature data of the construction inspection site, and perform noise reduction preprocessing on the three-dimensional feature data of the construction inspection site to generate quantitative three-dimensional feature data of the construction inspection site;

[0009] S2. Calling the quantified three-dimensional feature data of the construction inspection site, establishing a spatial rectangular coordinate system with the foundation of the construction inspection site as the base surface, dividing the three-dimensional features of the construction inspection site into spatial grid blocks using cubes of equal side lengths, and collecting the coordinates of the vertices of the cube spatial grid blocks in the spatial rectangular coordinate system to generate the three-dimensional coordinate data of the construction inspection site spatial grid;

[0010] S3. Matching the three-dimensional coordinate data of the construction inspection site spatial grid with the quantified three-dimensional feature data of the construction inspection site to display the relative spatial positions of the three-dimensional features of the construction inspection site and the three-dimensional coordinates of the construction inspection site spatial grid; based on the inspection location of the three-dimensional features of the construction site, selecting the three-dimensional coordinate points of the construction inspection site spatial grid adjacent to the inspection location and orderly collecting them to form the drone inspection operation path data;

[0011] S4. While the UAV is performing construction inspection according to the UAV inspection operation path data, the distance value between the UAV body and surrounding obstacles is collected in real time;

[0012] S5. Numerical analysis is performed on the distance between the drone body and the surrounding obstacles and the set safety distance value. When the distance between the drone body and the surrounding obstacles is equal to the safety distance value, it means that at this moment, the drone continues to move forward along the sensing side of the body and has a high probability of 80% to 100% contact and collision with the surrounding obstacles. At this time, the drone maintains a safe distance value from the obstacles on the sensing side of the body during the inspection operation path. When the distance between the drone body and the surrounding obstacles is less than the safety distance value, it means that at this moment, the drone continues to move forward along the sensing side of the body and has a high probability of 80% to 100% contact and collision with the surrounding obstacles. At this time, the drone first flies in the opposite direction along the sensing side of the body to increase the distance from the obstacles until it is equal to the safety distance value, and then continues to fly according to the inspection operation path. When the distance between the drone body and the surrounding obstacles is greater than the safety distance value, it means that at this moment, the drone continues to move forward along the sensing side of the body and has a low probability of 20% to 0% contact and collision with the surrounding obstacles. At this time, the drone continues to fly according to the inspection operation path.

[0013] S6. The drone flies to the construction inspection location according to the three-dimensional coordinate data of the construction inspection site space grid in the drone inspection operation path data, and collects characteristic image data of the construction inspection site;

[0014] S7. Use a recognition algorithm to perform image feature matching processing on the construction inspection site feature image data and the construction inspection site quality problem image data to identify the quality problem points at the construction inspection site; and record and collect the identified construction inspection site quality problem image data and the construction inspection site coordinates.

[0015] Preferably, the steps of scanning and collecting the three-dimensional features of the construction inspection site to generate the three-dimensional feature data of the construction inspection site, and performing noise reduction preprocessing on the three-dimensional feature data of the construction inspection site to generate the quantized three-dimensional feature data of the construction inspection site are as follows:

[0016] S11. The user operates a drone equipped with a laser scanner to scan the construction inspection site to generate three-dimensional feature data A of the construction inspection site;

[0017] S12. Use a Gaussian filtering algorithm to perform noise reduction preprocessing on the three-dimensional feature data A of the construction inspection site to generate quantitative three-dimensional feature data A′ of the construction inspection site.

[0018] Preferably, the steps of calling the quantified three-dimensional feature data of the construction inspection site, establishing a spatial rectangular coordinate system with the foundation of the construction inspection site as the base surface, dividing the three-dimensional features of the construction inspection site into spatial grid blocks using cubes of equal sides, and collecting the coordinates of the vertices of the cube spatial grid blocks in the spatial rectangular coordinate system to generate the three-dimensional coordinate data of the construction inspection site spatial grid are as follows:

[0019] S21, calling the quantitative construction inspection site three-dimensional feature data A';

[0020] S22, establishing a spatial rectangular coordinate system O using the construction inspection site foundation as a base surface;

[0021] S23, using cubes of equal length to divide the quantified construction inspection site three-dimensional feature data A′ into spatial grid blocks, collecting the coordinate values ​​of the cube space grid blocks’ vertices in the spatial rectangular coordinate system O to generate the construction inspection site spatial grid three-dimensional coordinate data set B = (b1,…,b n ,…,b r ), n=1,2,3,…,r; where b n represents the three-dimensional coordinate data of the spatial grid of the nth construction inspection site, r represents the maximum number of three-dimensional coordinates of the spatial grid of the construction inspection site, where b n =(x n ,y n ,z n ), x n Indicates the horizontal coordinate corresponding to the three-dimensional coordinate data of the spatial grid of the nth construction inspection site, y n Indicates the vertical coordinate corresponding to the three-dimensional coordinate data of the spatial grid of the nth construction inspection site, z n Indicates the vertical coordinate corresponding to the three-dimensional coordinate data of the spatial grid of the nth construction inspection site.

[0022] Preferably, the steps of matching the three-dimensional coordinate data of the construction inspection site space grid with the quantified three-dimensional feature data of the construction inspection site to display the relative spatial positions of the three-dimensional features of the construction inspection site and the three-dimensional coordinates of the construction inspection site space grid; and selecting the three-dimensional coordinate points of the construction inspection site space grid adjacent to the inspection position and orderly collecting them to form the drone inspection operation path data according to the inspection position of the three-dimensional features of the construction site are as follows:

[0023] S31, obtaining quantitative construction inspection site three-dimensional feature data A′ and construction inspection site spatial grid three-dimensional coordinate data B;

[0024] S32: Match and map the quantified three-dimensional feature data A′ of the construction inspection site with the three-dimensional coordinate data B of the spatial grid of the construction inspection site using the foundation of the construction inspection site and the spatial rectangular coordinate base O. Match and display the three-dimensional features of the construction inspection site with the three-dimensional coordinates of the spatial grid corresponding to the three-dimensional features of the construction inspection site.

[0025] S33. The operator selects the construction inspection site spatial grid three-dimensional coordinate data corresponding to the external adjacent space of the inspection location of the construction inspection site based on the displayed spatial grid three-dimensional coordinate data according to the three-dimensional feature matching of the construction inspection site. n , for the selected construction inspection site space grid three-dimensional coordinate data b n Generate the UAV inspection operation path data set C=(c1,…,c j ,…,c m ), j=1,2,3,…,m;1≤m≤r,where c j represents the jth UAV inspection operation path data, m represents the maximum number of UAV inspection operation path data, where c j =(x j ,y j ,z j ), x j Indicates the horizontal coordinate corresponding to the j-th UAV inspection operation path data, y j Indicates the vertical coordinate corresponding to the j-th UAV inspection operation path data, z j Indicates the vertical coordinate corresponding to the j-th UAV inspection operation path data.

[0026] Preferably, the operation steps of collecting the distance value between the drone body and surrounding obstacles in real time during the drone's construction inspection flight according to the drone inspection operation path data are as follows:

[0027] S41, the drone follows the drone inspection operation path data c in the drone inspection operation path data set C. j The coordinate points of the drone are used for orderly construction inspection flights. With the help of distance sensors evenly distributed on the surface of the drone, the distance values ​​between the drone body and the surrounding obstacles are measured online in real time, and a distance value set E between the drone body and the surrounding obstacles is established. t =(e 1t ,…,e vt ,…,e ut ), v=1,2,3,…,u, u represents the maximum value of the distance between the UAV body and the surrounding obstacles; where E t Represents the set of distance values ​​between the drone body and surrounding obstacles at time t, e vt The unit is meter, e vtIt represents the distance between the vth UAV body and the surrounding obstacles measured online by the distance sensor at time t.

[0028] Preferably, a numerical analysis is performed on the distance value between the drone body and the surrounding obstacles and the set safety distance value. When the distance value between the drone body and the surrounding obstacles is equal to the safety distance value, it means that at this moment, the drone continues to move forward along the sensing side of the body and has a high probability of 80% to 100% contact and collision with the surrounding obstacles. At this time, the drone maintains a safe distance value from the obstacles on the sensing side of the body during the flight process of the inspection operation path; when the distance value between the drone body and the surrounding obstacles is less than the safety distance value, it means that at this moment, the drone continues to move forward along the sensing side of the body and has a high probability of 80% to 100% contact and collision with the surrounding obstacles. At this time, the drone first flies in the opposite direction along the sensing side of the body to increase the distance from the obstacles until it is equal to the safety distance value, and then continues to fly according to the inspection operation path; when the distance value between the drone body and the surrounding obstacles is greater than the safety distance value, it means that at this moment, the drone continues to move forward along the sensing side of the body and has a low probability of 20% to 0% contact and collision with the surrounding obstacles. At this time, the operating steps for the drone to continue flying according to the inspection operation path are as follows:

[0029] S51, obtaining a set of distance values ​​E between the drone body and surrounding obstacles;

[0030] S52. Establish a safe distance value S between the drone and surrounding obstacles. The unit of the safe distance value S is meters. The setting of the safe distance value S needs to take into account the size of the drone and the complexity of the construction inspection site. The larger the size of the drone, the larger the safe distance value S. The more complex the construction inspection site, the larger the safe distance value S.

[0031] S53: Determine the distance value e between the drone body and surrounding obstacles vt The numerical value of the safety distance value S;

[0032] When e vt =S, it means that at time t, the drone continues to move forward along the sensing side of the aircraft and has an 80% to 100% probability of contact and collision with surrounding obstacles. At this time, the drone maintains a safe distance from the obstacles on the sensing side of the aircraft according to the flight process of the drone inspection operation path data set C;

[0033] When e vt <S, it means that at time t, the drone continues to move forward along the sensing side of the body and has an 80% to 100% probability of contact and collision with surrounding obstacles. At this time, the drone first flies in the opposite direction of the sensing side of the body to increase the distance between it and the obstacle until it reaches the safe distance value S, and then continues to fly according to the drone inspection operation path data set C;

[0034] When e vt ﹥S, it means that at time t, the drone continues to move forward along the sensing side of the body and collides with surrounding obstacles with a small probability value of 20% to 0%. At this time, the drone continues to fly according to the drone inspection operation path data set C.

[0035] Preferably, the operation steps of collecting characteristic image data of the construction inspection site after the drone flies to the construction inspection location according to the three-dimensional coordinate data of the construction inspection site space grid in the drone inspection operation path data are as follows:

[0036] S61, the UAV follows the UAV inspection operation path data c in the UAV inspection operation path data set C. j The corresponding construction inspection site spatial grid three-dimensional coordinate data is flown to the construction inspection location and the onboard camera is used to shoot the inspection location to generate construction inspection site feature image data.

[0037] Preferably, the steps of performing image feature matching processing on the construction inspection site feature image data and the construction inspection site quality problem image data using a recognition algorithm to identify quality problem points at the construction inspection site; and recording and collecting the identified construction inspection site quality problem image data and the construction inspection site coordinates are as follows:

[0038] S71. Acquire construction inspection site feature image data at the inspection location

[0039] S72, establish the construction inspection site quality problem image data matrix G = (g1, ..., g w ,…,g q ), w=1,2,3,…,q; where g w represents the wth type of construction inspection site quality problem image data, and q represents the maximum number of types of construction inspection site quality problems;

[0040] S73, using bee colony algorithm to transform the construction inspection site feature image data The construction inspection site quality problem image data g in the construction inspection site quality problem image data matrix G w Match and identify quality issues at the construction inspection site, including engineering equipment inspection, safety hazard investigation, construction quality inspection, and environmental protection inspection;

[0041] Engineering equipment inspection includes equipment operating status, equipment use and maintenance;

[0042] Safety hazard inspections include inspections of hazardous materials, temporary facilities, and firefighting equipment in the construction area;

[0043] Construction quality inspection includes inspection of construction building structure and construction progress;

[0044] Environmental protection inspections include sewage, dust, and garbage classification;

[0045] The bee colony algorithm is used to transform the construction inspection site feature image data Image data of quality problems on construction inspection site w The steps for identification and matching are as follows:

[0046] S731, initialize each honey source i, that is, the construction inspection site quality problem image data g in the construction inspection site quality problem image data matrix G w Set as honey source; set the number of honey sources to η, the current number of iterations t and the maximum number of iterations T; the initial position of honey source i is w i , the formula for generating the nectar source location is as follows: Randomly generate in the search space of the construction inspection site quality problem image data matrix G, where rand represents a random number in (0,1), and Λ represent the upper and lower bounds of the search space of the construction inspection site quality problem image data matrix G, respectively;

[0047] S732, the location of the nectar source w i Assign hired bees. At the beginning of the search, hired bees search the construction inspection site quality problem image data matrix G in the search space by all the construction inspection site quality problem image data g w The location of the nectar source w i Around, according to the formula w' i =w i +Ξ×(w i -w β ) Search and generate new nectar source location w' i , that is, traverse and search for all the construction inspection site quality problem image data g in the construction inspection site quality problem image data matrix G search space w and the construction inspection site feature image data Perform image feature matching; Indicates randomly selecting a nectar source that is not equal to i among n nectar sources, w β represents the position of the randomly selected honey source β in the search space of the construction inspection site quality problem image data matrix G, and Ξ represents a random number in the interval [-1, 1];

[0048] S733, based on the new nectar source location w′ iThe fitness of the construction inspection site is determined by the greedy selection method to determine the retained honey source, that is, the search space of the construction inspection site quality problem image data matrix G is searched for the construction inspection site feature image data. Matched construction inspection site quality problem image data g w ;

[0049] S734, the employed bees share the nectar source information, and the follower bees search the construction inspection site quality problem image data matrix G in the search space according to the probability value. Matched construction inspection site quality problem image data g w The bees follow the honey source and search for the construction inspection site feature image data shared by the employed bees. Matched construction inspection site quality problem image data g w The nectar source information is used to calculate the probability that the nectar source found by the employed bees will be followed according to the probability formula. where Z i Indicates the location of the new nectar source w' i The probability of being followed by an employed bee, Ω i The following bees indicate the location of the new nectar source w' shared by the employed bees i Adaptability;

[0050] S735. Follower bees search in the same way as employed bees and determine the retained nectar source i according to the greedy selection method, that is, they search for the nectar source i that matches the construction inspection site feature image data in the search space of the construction inspection site quality problem image data matrix G. Matched construction inspection site quality problem image data g w ;

[0051] S736, during the random search process of the scout bee in the search space of the construction inspection site quality problem image data matrix G, if the nectar source location w i After t iterations of searching, the iteration threshold is reached. No better nectar source was found. The nectar source location is w i The bees will be abandoned, and the corresponding employed bees will become scout bees. The calculation formula for the location of new nectar sources randomly generated by scout bees in the search space is as follows: in It represents the position of the new honey source i in the search space of the quality problem image data matrix G of the construction inspection site after t+1 iterations, It represents the position of the new honey source i generated after t iterations in the search space of the quality problem image data matrix G of the construction inspection site;

[0052] S737: Determine whether the algorithm meets the maximum search iteration number T condition. If so, terminate and output the construction inspection site feature image data. The construction inspection site quality problem image data g in the construction inspection site quality problem image data matrix G w If the result is a match, go to step S732;

[0053] S74, when the construction inspection site feature image data Image data of quality problems on construction inspection site w If the match is successful, it means that the UAV inspection operation path data c j There are quality problem points at the corresponding construction inspection locations, and the drone inspection operation path data c is recorded at the inspection locations j and construction inspection site feature image data

[0054] When the construction inspection site feature image data Image data of quality problems on construction inspection site w If the match is not successful, it means that the UAV inspection operation path data c j If there are no quality problem points at the corresponding construction inspection location, the drone will continue to fly and perform the inspection according to the drone inspection operation path data set C until the inspection operation is completed.

[0055] A system for implementing the above-mentioned method for engineering construction inspection, comprising an engineering construction inspection site data acquisition module, an unmanned aerial vehicle inspection path planning module, an unmanned aerial vehicle inspection obstacle avoidance management module, and an engineering construction inspection quality problem point data collection module;

[0056] The engineering construction inspection site data acquisition module includes a construction inspection site 3D scene feature acquisition unit and a construction inspection site 3D scene feature processing unit;

[0057] The construction inspection site 3D scene feature acquisition unit is a human-operated drone equipped with a laser scanner to scan and collect 3D features of the construction inspection site to generate 3D feature data of the construction inspection site; the construction inspection site 3D scene feature processing unit is a Gaussian filter algorithm for noise reduction pre-processing of the 3D feature data of the construction inspection site to generate quantized 3D feature data of the construction inspection site;

[0058] The UAV inspection path planning module includes a construction inspection site three-dimensional scene feature data calling unit, a construction inspection site three-dimensional scene feature data space grid coordinate establishment unit, and a UAV inspection operation path formulation unit;

[0059] The three-dimensional scene feature data calling unit of the construction inspection site calls the quantitative three-dimensional feature data of the construction inspection site; the three-dimensional scene feature data spatial grid coordinate establishing unit of the construction inspection site establishes a spatial rectangular coordinate system with the foundation of the construction inspection site as the base surface, divides the three-dimensional features of the construction inspection site into spatial grid blocks using cubes of equal sides, collects the coordinates of the vertices of the cube spatial grid blocks in the spatial rectangular coordinate system to generate the three-dimensional coordinate data of the construction inspection site spatial grid; the unmanned aerial vehicle inspection operation path formulation unit matches the three-dimensional coordinate data of the construction inspection site spatial grid with the quantitative three-dimensional feature data of the construction inspection site to display the relative spatial position of the three-dimensional features of the construction inspection site and the three-dimensional coordinates of the three-dimensional grid of the construction inspection site; according to the inspection position of the three-dimensional features of the construction site, the three-dimensional coordinate points of the spatial grid of the construction inspection site adjacent to the inspection position are selected and orderly collected to form the unmanned aerial vehicle inspection operation path data;

[0060] The UAV inspection and obstacle avoidance management module includes a UAV and obstacle distance value collection unit, a UAV and obstacle safety distance value setting unit, and a UAV and obstacle avoidance analysis and early warning unit;

[0061] The UAV and obstacle distance value acquisition unit uses a distance measuring sensor to collect the distance value between the UAV body and surrounding obstacles in real time during the UAV construction inspection flight process; the UAV and obstacle safety distance value setting unit is used to set the safety distance value; the UAV and obstacle avoidance analysis and early warning unit is used to perform numerical analysis on the distance value between the UAV body and surrounding obstacles and the set safety distance value. When the distance value between the UAV body and the surrounding obstacles is equal to the safety distance value, the UAV maintains a safe distance value from the obstacles on the sensing side of the body during the inspection operation path; when the distance value between the UAV body and the surrounding obstacles is less than the safety distance value, the UAV first flies in the opposite direction of the sensing side of the body to increase the distance from the obstacles until it is equal to the safety distance value and then continues to fly according to the inspection operation path; when the distance value between the UAV body and the surrounding obstacles is greater than the safety distance value, the UAV continues to fly according to the inspection operation path;

[0062] The engineering construction inspection quality problem point data collection module includes a construction inspection site feature image data acquisition unit, a construction inspection site quality problem image data storage unit, and a construction inspection site quality problem identification and coordinate data recording unit;

[0063] The construction inspection site characteristic image data acquisition unit uses a drone equipped with a shooting lens to collect the construction inspection site characteristic image data of the construction inspection location; the construction inspection site quality problem image data storage unit is used to store the construction inspection site quality problem image data; the construction inspection site quality problem identification and coordinate data recording unit is used to use a bee swarm algorithm to match the construction inspection site characteristic image data with the construction inspection site quality problem image data to perform image feature matching processing to identify quality problem points at the construction inspection site; and record and collect the identified construction inspection site quality problem image data and the construction inspection site coordinates.

[0064] (3) Beneficial effects

[0065] The present invention provides a method and system for engineering construction inspection, which has the following beneficial effects:

[0066] 1. Through the construction inspection site 3D scene feature acquisition unit, the drone equipped with a laser scanner is manually operated to scan and collect the 3D features of the construction inspection site to generate the 3D feature data of the construction inspection site. With the help of the drone and the laser scanner, the spatial 3D model data of the construction inspection site is accurately and quickly established. The construction inspection site 3D scene feature data spatial grid coordinate establishment unit and the drone inspection operation path formulation unit cooperate to establish a spatial rectangular coordinate system, divide the 3D features of the construction inspection site into spatial grid blocks, obtain the coordinate data of the grid block vertices in the spatial rectangular coordinate system and match them with the spatial position of the construction inspection site 3D model, scientifically establish the spatial 3D model coordinate data of the construction inspection, provide coordinate support for the drone inspection, manually select the inspection position of the construction inspection site, and orderly gather the grid coordinates in the spatial rectangular coordinate system to generate the drone inspection operation path data. The drone construction inspection path is flexibly planned according to the inspection position, thereby improving the efficiency and quality of the construction inspection.

[0067] 2. The UAV and obstacle distance value acquisition unit uses the distance measurement sensor to collect the distance value between the UAV body and surrounding obstacles in real time during the UAV construction inspection flight, so as to achieve accurate obstacle avoidance during the UAV construction inspection flight; the UAV and obstacle avoidance analysis and warning unit uses the three judgment results of equal, less than, and greater than the distance value between the UAV body and the surrounding obstacles and the safety distance value to respectively use the UAV to maintain a safe distance value from the obstacles on the sensing side of the body during the inspection operation path; the UAV first flies in the opposite direction of the sensing side of the body to increase the distance between the obstacle and the body until it is equal to the safety distance value and then continues to fly according to the inspection operation path; the UAV continues to fly along the inspection operation path. Three different obstacle avoidance schemes control the UAV to accurately avoid obstacles, which improves the reliability of the UAV in engineering construction inspection and effectively ensures the safety of engineering construction.

[0068] 3. The construction inspection site feature image data acquisition unit uses a drone equipped with a camera to collect the construction inspection site feature image data of the construction inspection location, so as to accurately obtain the latest construction status of the construction inspection site and ensure the accuracy of the construction inspection results; the construction inspection site quality problem identification and coordinate data recording unit uses a bee swarm algorithm to match the construction inspection site feature image data with the construction inspection site quality problem image data to identify the quality problem points at the construction inspection site; when quality problem points are identified at the construction inspection site, the identified construction inspection site quality problem image data and the inspection location spatial coordinates of the construction inspection site are recorded and collected, so as to facilitate operators to quickly obtain the quality problem points and coordinates of the construction site, thereby improving the efficiency and accuracy of identifying quality problem points in the construction inspection operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] Figure 1 This is a module composition diagram of a construction inspection system provided by the present invention;

[0070] Figure 2 for Figure 1 The figure shows a flow chart of a method for engineering construction inspection. DETAILED DESCRIPTION

[0071] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0072] The embodiments of the method and system for engineering construction inspection are as follows:

[0073] See also Figure 1 - Figure 2 A method for engineering construction inspection, the method comprising the following steps:

[0074] S1. Scan and collect three-dimensional features of the construction inspection site to generate three-dimensional feature data of the construction inspection site, and perform noise reduction preprocessing on the three-dimensional feature data of the construction inspection site to generate quantitative three-dimensional feature data of the construction inspection site;

[0075] S2. Calling the quantified three-dimensional feature data of the construction inspection site, establishing a spatial rectangular coordinate system with the foundation of the construction inspection site as the base surface, dividing the three-dimensional features of the construction inspection site into spatial grid blocks using cubes of equal side lengths, and collecting the coordinates of the vertices of the cube spatial grid blocks in the spatial rectangular coordinate system to generate the three-dimensional coordinate data of the construction inspection site spatial grid;

[0076] S3. Matching the three-dimensional coordinate data of the construction inspection site spatial grid with the quantified three-dimensional feature data of the construction inspection site to display the relative spatial positions of the three-dimensional features of the construction inspection site and the three-dimensional coordinates of the construction inspection site spatial grid; based on the inspection location of the three-dimensional features of the construction site, selecting the three-dimensional coordinate points of the construction inspection site spatial grid adjacent to the inspection location and orderly collecting them to form the drone inspection operation path data;

[0077] S4. While the UAV is performing construction inspection according to the UAV inspection operation path data, the distance value between the UAV body and surrounding obstacles is collected in real time;

[0078] S5. Numerical analysis is performed on the distance between the drone body and the surrounding obstacles and the set safety distance value. When the distance between the drone body and the surrounding obstacles is equal to the safety distance value, it means that at this moment, the drone continues to move forward along the sensing side of the body and has a high probability of 80% to 100% contact and collision with the surrounding obstacles. At this time, the drone maintains a safe distance value from the obstacles on the sensing side of the body during the inspection operation path. When the distance between the drone body and the surrounding obstacles is less than the safety distance value, it means that at this moment, the drone continues to move forward along the sensing side of the body and has a high probability of 80% to 100% contact and collision with the surrounding obstacles. At this time, the drone first flies in the opposite direction along the sensing side of the body to increase the distance from the obstacles until it is equal to the safety distance value, and then continues to fly according to the inspection operation path. When the distance between the drone body and the surrounding obstacles is greater than the safety distance value, it means that at this moment, the drone continues to move forward along the sensing side of the body and has a low probability of 20% to 0% contact and collision with the surrounding obstacles. At this time, the drone continues to fly according to the inspection operation path.

[0079] S6. The drone flies to the construction inspection location according to the three-dimensional coordinate data of the construction inspection site space grid in the drone inspection operation path data, and collects characteristic image data of the construction inspection site;

[0080] S7. Use a recognition algorithm to perform image feature matching processing on the construction inspection site feature image data and the construction inspection site quality problem image data to identify the quality problem points at the construction inspection site; and record and collect the identified construction inspection site quality problem image data and the construction inspection site coordinates.

[0081] For further information, see Figure 1 - Figure 2 The steps for scanning and collecting the three-dimensional features of the construction inspection site to generate the three-dimensional feature data of the construction inspection site and performing noise reduction preprocessing on the three-dimensional feature data of the construction inspection site to generate the quantitative three-dimensional feature data of the construction inspection site are as follows:

[0082] S11. The user operates a drone equipped with a laser scanner to scan the construction inspection site to generate three-dimensional feature data A of the construction inspection site;

[0083] S12. Use a Gaussian filtering algorithm to perform noise reduction preprocessing on the three-dimensional feature data A of the construction inspection site to generate quantitative three-dimensional feature data A′ of the construction inspection site.

[0084] The steps for calling the quantified 3D feature data of the construction inspection site and establishing a spatial rectangular coordinate system based on the foundation of the construction inspection site are as follows:

[0085] S21, calling the quantitative construction inspection site three-dimensional feature data A';

[0086] S22, establishing a spatial rectangular coordinate system O using the construction inspection site foundation as a base surface;

[0087] S23, using cubes of equal length to divide the quantified construction inspection site three-dimensional feature data A′ into spatial grid blocks, collecting the coordinate values ​​of the cube space grid blocks’ vertices in the spatial rectangular coordinate system O to generate the construction inspection site spatial grid three-dimensional coordinate data set B = (b1,…,b n ,…,b r ), n=1,2,3,…,r; where b n represents the three-dimensional coordinate data of the spatial grid of the nth construction inspection site, r represents the maximum number of three-dimensional coordinates of the spatial grid of the construction inspection site, where b n =(x n ,y n ,z n ), x n Indicates the horizontal coordinate corresponding to the three-dimensional coordinate data of the spatial grid of the nth construction inspection site, y n Indicates the vertical coordinate corresponding to the three-dimensional coordinate data of the spatial grid of the nth construction inspection site, z n Indicates the vertical coordinate corresponding to the three-dimensional coordinate data of the spatial grid of the nth construction inspection site.

[0088] The three-dimensional coordinate data of the construction inspection site spatial grid is matched with the quantitative three-dimensional feature data of the construction inspection site to display the relative spatial position of the three-dimensional features of the construction inspection site and the three-dimensional coordinates of the construction inspection site spatial grid. Based on the inspection location of the three-dimensional features of the construction site, the three-dimensional coordinate points of the construction inspection site spatial grid adjacent to the inspection location are selected and orderly aggregated to form the drone inspection operation path data. The following steps are performed:

[0089] S31, obtaining quantitative construction inspection site three-dimensional feature data A′ and construction inspection site spatial grid three-dimensional coordinate data B;

[0090] S32: Match and map the quantified three-dimensional feature data A′ of the construction inspection site with the three-dimensional coordinate data B of the spatial grid of the construction inspection site using the foundation of the construction inspection site and the spatial rectangular coordinate base O. Match and display the three-dimensional features of the construction inspection site with the three-dimensional coordinates of the spatial grid corresponding to the three-dimensional features of the construction inspection site.

[0091] S33. The operator selects the construction inspection site spatial grid three-dimensional coordinate data corresponding to the external adjacent space of the inspection location of the construction inspection site based on the displayed spatial grid three-dimensional coordinate data according to the three-dimensional feature matching of the construction inspection site. n , for the selected construction inspection site space grid three-dimensional coordinate data b n Generate the UAV inspection operation path data set C=(c1,…,c j ,…,c m ), j=1,2,3,…,m;1≤m≤r,where c j represents the jth UAV inspection operation path data, m represents the maximum number of UAV inspection operation path data, where c j =(x j ,y j ,z j ), x j Indicates the horizontal coordinate corresponding to the j-th UAV inspection operation path data, y j Indicates the vertical coordinate corresponding to the j-th UAV inspection operation path data, z j Indicates the vertical coordinate corresponding to the j-th UAV inspection operation path data.

[0092] The 3D scene feature acquisition unit of the construction inspection site manually operates the drone equipped with a laser scanner to scan and collect the 3D features of the construction inspection site to generate the 3D feature data of the construction inspection site, and uses the drone and laser scanner to accurately and quickly establish the spatial 3D model data of the construction inspection site; the 3D scene feature data spatial grid coordinate establishment unit of the construction inspection site and the drone inspection operation path formulation unit cooperate to establish a spatial rectangular coordinate system, divide the 3D features of the construction inspection site into spatial grid blocks, obtain the coordinate data of the grid block vertices in the spatial rectangular coordinate system and match them with the spatial position of the 3D model of the construction inspection site, scientifically establish the spatial 3D model coordinate data of the construction inspection, provide coordinate support for the drone inspection, manually select the inspection position of the construction inspection site, and orderly gather the grid coordinates in the spatial rectangular coordinate system to generate the drone inspection operation path data, flexibly plan the drone construction inspection path according to the inspection position, and improve the efficiency and quality of the construction inspection.

[0093] For further information, see Figure 1 - Figure 2 The steps for collecting the distance between the drone body and surrounding obstacles in real time during the drone's construction inspection flight according to the drone inspection operation path data are as follows:

[0094] S41, the drone follows the drone inspection operation path data c in the drone inspection operation path data set C. j The coordinate points of the drone are used for orderly construction inspection flights. With the help of distance sensors evenly distributed on the surface of the drone, the distance values ​​between the drone body and the surrounding obstacles are measured online in real time, and a distance value set E between the drone body and the surrounding obstacles is established. t =(e 1t ,…,e vt ,…,e ut ), v=1,2,3,…,u, u represents the maximum value of the distance between the UAV body and the surrounding obstacles; where E t Represents the set of distance values ​​between the drone body and surrounding obstacles at time t, e vt The unit is meter, e vt It represents the distance between the vth UAV body and the surrounding obstacles measured online by the distance sensor at time t.

[0095] Numerical analysis is performed on the distance between the drone body and surrounding obstacles and the set safety distance value. When the distance between the drone body and the surrounding obstacles is equal to the safety distance value, it means that at this moment, the drone continues to move forward along the sensing side of the body and has a high probability of contact and collision with the surrounding obstacles. At this time, the drone maintains a safe distance value from the obstacles on the sensing side of the body during the inspection operation path. When the distance between the drone body and the surrounding obstacles is less than the safety distance value, it means that at this moment, the drone continues to move forward along the sensing side of the body and has a high probability of contact and collision with the surrounding obstacles. At this time, the drone first flies in the opposite direction of the sensing side of the body to increase the distance from the obstacle until it equals the safety distance value, and then continues to fly according to the inspection operation path. When the distance between the drone body and the surrounding obstacles is greater than the safety distance value, it means that at this moment, the drone continues to move forward along the sensing side of the body and has a low probability of contact and collision with the surrounding obstacles. At this time, the operation steps for the drone to continue flying according to the inspection operation path are as follows:

[0096] S51, obtaining a set of distance values ​​E between the drone body and surrounding obstacles;

[0097] S52. Establish a safe distance value S between the drone and surrounding obstacles. The unit of the safe distance value S is meters. The setting of the safe distance value S needs to take into account the size of the drone and the complexity of the construction inspection site. The larger the size of the drone, the larger the safe distance value S. The more complex the construction inspection site, the larger the safe distance value S.

[0098] S53: Determine the distance value e between the drone body and surrounding obstacles vt The numerical value of the safety distance value S;

[0099] When e vt =S, it means that at time t, the drone continues to move forward along the sensing side of the aircraft and has an 80% to 100% probability of contact and collision with surrounding obstacles. At this time, the drone maintains a safe distance from the obstacles on the sensing side of the aircraft according to the flight process of the drone inspection operation path data set C;

[0100] When e vt <S, it means that at time t, the drone continues to move forward along the sensing side of the body and has an 80% to 100% probability of contact and collision with surrounding obstacles. At this time, the drone first flies in the opposite direction of the sensing side of the body to increase the distance between it and the obstacle until it reaches the safe distance value S, and then continues to fly according to the drone inspection operation path data set C;

[0101] When e vt ﹥S, it means that at time t, the drone continues to move forward along the sensing side of the body and collides with surrounding obstacles with a small probability value of 20% to 0%. At this time, the drone continues to fly according to the drone inspection operation path data set C.

[0102] The drone and obstacle distance value acquisition unit uses the distance measurement sensor to collect the distance value between the drone body and surrounding obstacles in real time during the drone construction inspection flight, thereby realizing accurate obstacle avoidance during the drone construction inspection flight; the drone and obstacle avoidance analysis and early warning unit uses the three judgment results of equal, less than, and greater than the distance value between the drone body and surrounding obstacles and the safety distance value to respectively use the drone to maintain a safe distance value from the obstacles on the sensing side of the body during the inspection operation path; the drone first flies in the opposite direction of the sensing side of the body to increase the distance to the obstacle until it is equal to the safety distance value and then continues to fly according to the inspection operation path; the drone continues to fly according to the inspection operation path. The three different obstacle avoidance schemes control the drone to accurately avoid obstacles, thereby improving the reliability of drones in engineering construction inspections and effectively ensuring the safety of engineering construction.

[0103] For further information, see Figure 1 - Figure 2The operation steps for collecting characteristic image data of the construction inspection site after the drone flies to the construction inspection location according to the three-dimensional coordinate data of the construction inspection site space grid in the drone inspection operation path data are as follows:

[0104] S61, the UAV follows the UAV inspection operation path data c in the UAV inspection operation path data set C. j The corresponding construction inspection site spatial grid three-dimensional coordinate data is flown to the construction inspection location and the onboard camera is used to shoot the inspection location to generate construction inspection site feature image data.

[0105] The construction inspection site feature image data is matched with the construction inspection site quality problem image data using a recognition algorithm to identify the quality problem points at the construction inspection site. The steps for recording and collecting the identified construction inspection site quality problem image data and the construction inspection site coordinates are as follows:

[0106] S71. Acquire construction inspection site feature image data at the inspection location

[0107] S72, establish the construction inspection site quality problem image data matrix G = (g1, ..., g w ,…,g q ), w=1,2,3,…,q; where g w represents the wth type of construction inspection site quality problem image data, and q represents the maximum number of types of construction inspection site quality problems;

[0108] S73, using bee colony algorithm to transform the construction inspection site feature image data The construction inspection site quality problem image data g in the construction inspection site quality problem image data matrix G w Match and identify quality issues at the construction inspection site, including engineering equipment inspection, safety hazard investigation, construction quality inspection, and environmental protection inspection;

[0109] Engineering equipment inspection includes equipment operating status, equipment use and maintenance;

[0110] Safety hazard inspections include inspections of hazardous materials, temporary facilities, and firefighting equipment in the construction area;

[0111] Construction quality inspection includes inspection of construction building structure and construction progress;

[0112] Environmental protection inspections include sewage, dust, and garbage classification;

[0113] The bee colony algorithm is used to transform the construction inspection site feature image data Image data of quality problems on construction inspection site w The steps for identification and matching are as follows:

[0114] S731, initialize each honey source i, that is, the construction inspection site quality problem image data g in the construction inspection site quality problem image data matrix G w Set as honey source; set the number of honey sources to η, the current number of iterations t and the maximum number of iterations T; the initial position of honey source i is w i , the formula for generating the nectar source location is as follows: Randomly generate in the search space of the quality problem image data matrix G of the construction inspection site, where rand represents a random number in (0,1). and Λ represent the upper and lower bounds of the search space of the image data matrix G of quality problems at the construction inspection site, respectively;

[0115] S732, the location of the nectar source w i Assign hired bees. At the beginning of the search, hired bees search the construction inspection site quality problem image data matrix G in the search space consisting of all the construction inspection site quality problem image data g w The location of the nectar source w i Around, according to the formula w' i =w i +Ξ×(w i -w β ) Search and generate new nectar source location w' i , that is, traverse and search all the construction inspection site quality problem image data g in the construction inspection site quality problem image data matrix G search space w and construction inspection site feature image data Perform image feature matching; Indicates randomly selecting a nectar source that is not equal to i among n nectar sources, w β represents the position of the randomly selected honey source β in the search space of the quality problem image data matrix G at the construction inspection site, and Ξ represents a random number in the interval [-1,1];

[0116] S733, based on the new nectar source location w′ i The fitness of the nectar source is determined by the greedy selection method, that is, the nectar source that is retained is searched for in the search space of the construction inspection site quality problem image data matrix G. Matched construction inspection site quality problem image data g w ;

[0117] S734, the employed bees share the nectar source information, and the follower bees search the construction inspection site quality problem image data matrix G in the search space according to the probability value. Matched construction inspection site quality problem image data g w Search for nectar sources, and follow the bees based on the characteristic image data of the construction inspection site shared by the employed bees Matched construction inspection site quality problem image data g w The nectar source information is used to calculate the probability that the nectar source found by the employed bees will be followed according to the probability formula. where Z i Indicates the location of the new nectar source w' i The probability of being followed by an employed bee, Ω i The following bees indicate the location of the new nectar source w' shared by the employed bees i Adaptability;

[0118] S735. Follower bees search in the same way as employed bees and determine the nectar source i to be retained based on the greedy selection method, that is, they search for the nectar source i that matches the construction inspection site feature image data in the search space of the construction inspection site quality problem image data matrix G. Matched construction inspection site quality problem image data g w ;

[0119] S736, during the random search process of the scout bee in the search space of the quality problem image data matrix G at the construction inspection site, if the nectar source location w i After t iterations of searching, the iteration threshold is reached. No better nectar source was found. The nectar source location is w i The bees will be abandoned, and the corresponding employed bees will become scout bees. The calculation formula for the location of new nectar sources randomly generated by scout bees in the search space is as follows: in It represents the position of the new honey source i in the search space of the quality problem image data matrix G of the construction inspection site after t+1 iterations, It represents the position of the new nectar source i in the search space of the quality problem image data matrix G of the construction inspection site after t iterations;

[0120] S737: Determine whether the algorithm meets the maximum search iteration number T condition. If so, terminate and output the construction inspection site feature image data. The construction inspection site quality problem image data g in the construction inspection site quality problem image data matrix G w If the result is a match, go to step S732;

[0121] S74, when the construction inspection site feature image data Image data of quality problems on construction inspection site w If the match is successful, it means that the UAV inspection operation path data c jThere are quality problem points at the corresponding construction inspection locations, and the drone inspection operation path data c is recorded at the inspection locations j and construction inspection site feature image data

[0122] When the construction inspection site feature image data Image data of quality problems on construction inspection site w If the match is not successful, it means that the UAV inspection operation path data c j If there are no quality problem points at the corresponding construction inspection location, the drone will continue to fly and perform the inspection according to the drone inspection operation path data set C until the inspection operation is completed.

[0123] The construction inspection site characteristic image data acquisition unit uses a drone equipped with a shooting lens to collect the construction inspection site characteristic image data of the construction inspection location, so as to accurately obtain the latest construction status of the construction inspection site and ensure the accuracy of the construction inspection results; the construction inspection site quality problem identification and coordinate data recording unit uses a bee swarm algorithm to match the construction inspection site characteristic image data with the construction inspection site quality problem image data to identify the quality problem points at the construction inspection site; when quality problem points are identified at the construction inspection site, the identified construction inspection site quality problem image data and the inspection location spatial coordinates of the construction inspection site are recorded and collected, so as to facilitate operators to quickly obtain the quality problem points and coordinates of the construction site, thereby improving the efficiency and accuracy of identifying quality problem points in the construction inspection operation.

[0124] A system for implementing a method for engineering construction inspection, the system includes an engineering construction inspection site data acquisition module, an unmanned aerial vehicle inspection path planning module, an unmanned aerial vehicle inspection obstacle avoidance management module, and an engineering construction inspection quality problem point data collection module;

[0125] The construction inspection site data acquisition module includes a construction inspection site 3D scene feature acquisition unit and a construction inspection site 3D scene feature processing unit;

[0126] The construction inspection site 3D scene feature acquisition unit uses a human-operated drone equipped with a laser scanner to scan and collect 3D features of the construction inspection site to generate 3D feature data of the construction inspection site. The construction inspection site 3D scene feature processing unit uses a Gaussian filter algorithm to perform noise reduction pre-processing on the 3D feature data of the construction inspection site to generate quantitative 3D feature data of the construction inspection site.

[0127] The UAV inspection path planning module includes a construction inspection site 3D scene feature data calling unit, a construction inspection site 3D scene feature data spatial grid coordinate establishment unit, and a UAV inspection operation path formulation unit;

[0128] A construction inspection site three-dimensional scene feature data calling unit calls the quantitative construction inspection site three-dimensional feature data; a construction inspection site three-dimensional scene feature data spatial grid coordinate establishment unit establishes a spatial rectangular coordinate system based on the construction inspection site foundation as the base surface, divides the three-dimensional features of the construction inspection site into spatial grid blocks using cubes of equal sides, collects the coordinates of the vertices of the cube spatial grid blocks in the spatial rectangular coordinate system to generate the construction inspection site spatial grid three-dimensional coordinate data; a drone inspection operation path formulation unit matches the construction inspection site spatial grid three-dimensional coordinate data with the quantitative construction inspection site three-dimensional feature data to display the relative spatial position of the construction inspection site three-dimensional features and the construction inspection site spatial grid three-dimensional coordinates; according to the inspection position of the construction site three-dimensional features, the construction inspection site spatial grid three-dimensional coordinate points adjacent to the inspection position are selected and orderly aggregated to form the drone inspection operation path data;

[0129] The UAV inspection and obstacle avoidance management module includes a UAV and obstacle distance value collection unit, a UAV and obstacle safety distance value setting unit, and a UAV and obstacle avoidance analysis and early warning unit;

[0130] The drone and obstacle distance value acquisition unit uses a distance measurement sensor to collect the distance value between the drone body and surrounding obstacles in real time during the drone construction inspection flight process; the drone and obstacle safety distance value setting unit is used to set the safety distance value; the drone and obstacle avoidance analysis and early warning unit is used to perform numerical analysis on the distance value between the drone body and surrounding obstacles and the set safety distance value. When the distance value between the drone body and the surrounding obstacles is equal to the safety distance value, it means that at this moment the drone continues to move forward along the sensing side of the body and has an 80% to 100% probability of contact and collision with the surrounding obstacles. At this time, the drone flies along the inspection operation path and Maintain a safe distance value for flying with obstacles on the sensing side; when the distance between the drone body and the surrounding obstacles is less than the safe distance value, it means that at this moment the drone continues to move forward along the sensing side of the body and there is a high probability of 80% to 100% contact and collision with the surrounding obstacles. At this time, the drone first flies in the opposite direction of the sensing side of the body to increase the distance from the obstacle until it equals the safe distance value, and then continues to fly according to the inspection operation path; when the distance between the drone body and the surrounding obstacles is greater than the safe distance value, it means that at this moment the drone continues to move forward along the sensing side of the body and there is a low probability of 20% to 0% contact and collision with the surrounding obstacles. At this time, the drone continues to fly according to the inspection operation path;

[0131] The data collection module for engineering construction inspection quality problem points includes a construction inspection site feature image data acquisition unit, a construction inspection site quality problem image data storage unit, and a construction inspection site quality problem identification and coordinate data recording unit;

[0132] The construction inspection site characteristic image data acquisition unit uses a drone equipped with a shooting lens to collect the construction inspection site characteristic image data of the construction inspection location; the construction inspection site quality problem image data storage unit is used to store the construction inspection site quality problem image data; the construction inspection site quality problem identification and coordinate data recording unit is used to use a bee swarm algorithm to match the construction inspection site characteristic image data with the construction inspection site quality problem image data to perform image feature matching processing to identify quality problem points at the construction inspection site; and record and collect the identified construction inspection site quality problem image data and the construction inspection site coordinates.

[0133] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A method for engineering construction inspection, characterized in that: The method comprises the following steps: S1. Scan and collect the three-dimensional features of the construction inspection site to generate three-dimensional feature data of the construction inspection site, and perform noise reduction preprocessing on the three-dimensional feature data of the construction inspection site to generate quantitative three-dimensional feature data of the construction inspection site; S2. Call the quantitative three-dimensional feature data of the construction inspection site, establish a spatial rectangular coordinate system with the foundation of the construction inspection site as the base surface, divide the three-dimensional features of the construction inspection site into spatial grid blocks using cubes of equal length, collect the coordinates of the vertices of the cube spatial grid blocks in the spatial rectangular coordinate system to generate the three-dimensional coordinate data of the construction inspection site spatial grid; S3, matching the three-dimensional coordinate data of the construction inspection site space grid with the quantified three-dimensional feature data of the construction inspection site to display the relative spatial position of the three-dimensional features of the construction inspection site and the three-dimensional coordinates of the construction inspection site space grid; according to the inspection position of the three-dimensional features of the construction site, selecting the three-dimensional coordinate points of the construction inspection site space grid adjacent to the inspection position and orderly collecting them to form the UAV inspection operation path data; S4. During the construction inspection flight of the UAV according to the inspection operation path data of the UAV, the distance value between the UAV body and the surrounding obstacles is collected in real time; S5. Numerical analysis is performed on the distance between the drone body and surrounding obstacles and the set safety distance value. When the distance between the drone body and surrounding obstacles is equal to the safety distance value, the drone flies along the inspection operation path and maintains a safety distance value from the obstacles on the sensing side of the drone body; when the distance between the drone body and surrounding obstacles is less than the safety distance value, the drone first flies in the opposite direction of the sensing side of the drone body to increase the distance from the obstacles until it is equal to the safety distance value and then continues to fly along the inspection operation path; when the distance between the drone body and surrounding obstacles is greater than the safety distance value, the drone continues to fly along the inspection operation path; S6. The drone flies to the construction inspection location according to the three-dimensional coordinate data of the construction inspection site space grid in the drone inspection operation path data, and then collects characteristic image data of the construction inspection site; S7. Use a recognition algorithm to perform image feature matching processing on the construction inspection site feature image data and the construction inspection site quality problem image data to identify the quality problem points at the construction inspection site; and record and collect the identified construction inspection site quality problem image data and the construction inspection site coordinates.

2. A method for engineering construction inspection according to claim 1, characterized in that: The S1 comprises the following steps: S11, the user operates the drone equipped with a laser scanner to scan the construction inspection site to generate three-dimensional feature data A of the construction inspection site; S12. Use a Gaussian filtering algorithm to perform noise reduction preprocessing on the three-dimensional feature data A of the construction inspection site to generate quantitative three-dimensional feature data A′ of the construction inspection site.

3. A method for engineering construction inspection according to claim 2, characterized in that: The S2 comprises the following steps: S21, calling the quantitative construction inspection site three-dimensional feature data A′; S22, establishing a spatial rectangular coordinate system O using the construction inspection site foundation as a base surface; S23, using cubes of equal length to divide the quantified construction inspection site three-dimensional feature data A′ into spatial grid blocks, collecting the coordinate values ​​of the cube space grid blocks’ vertices in the spatial rectangular coordinate system O to generate the construction inspection site spatial grid three-dimensional coordinate data set B = (b1,…,b n ,…,b r ), n=1,2,3,…,r; where b n represents the three-dimensional coordinate data of the nth construction inspection site spatial grid, and r represents the maximum number of three-dimensional coordinates of the construction inspection site spatial grid.

4. A method for engineering construction inspection according to claim 3, characterized in that: The S3 comprises the following steps: S31, obtaining quantitative construction inspection site three-dimensional feature data A′ and construction inspection site spatial grid three-dimensional coordinate data B; S32, the quantified construction inspection site three-dimensional feature data A′ and the construction inspection site spatial grid three-dimensional coordinate data B are matched and mapped with the construction inspection site foundation and the spatial rectangular coordinate O base surface, and the three-dimensional features of the construction inspection site and the three-dimensional coordinates of the spatial grid corresponding to the three-dimensional features of the construction inspection site are matched and displayed in spatial position; S33, the operator selects the construction inspection site spatial grid three-dimensional coordinate data b corresponding to the external adjacent space of the construction inspection site inspection position according to the displayed spatial grid three-dimensional coordinate data of the construction inspection site three-dimensional feature matching n , for the selected construction inspection site spatial grid three-dimensional coordinate data b n Generate the UAV inspection operation path data set C=(c1,…,c j ,…,c m ), j = 1, 2, 3, ..., m; 1 ≤ m ≤ r, where c j represents the j-th UAV inspection operation path data, and m represents the maximum number of UAV inspection operation path data.

5. A method for engineering construction inspection according to claim 4, characterized in that: The S4 comprises the following steps: S41. The UAV performs construction inspection flights in an orderly manner according to the coordinate points of the UAV inspection operation path data cj in the UAV inspection operation path data set C. The distance values ​​between the UAV body and the surrounding obstacles are measured online in real time by using the distance sensors evenly distributed on the surface of the UAV, and a distance value set E between the UAV body and the surrounding obstacles is established. t =(e 1t ,…,e vt ,…,e ut ), v = 1, 2, 3, ..., u, u represents the maximum value of the distance between the drone body and the surrounding obstacles; where E t represents the distance value set between the drone body and surrounding obstacles at time t, e vt The unit is meter, e vt It represents the distance value between the vth UAV body and the surrounding obstacles measured online by the distance sensor at time t.

6. A method for engineering construction inspection according to claim 5, characterized in that: The S5 comprises the following steps: S51, obtaining a distance value set E between the drone body and surrounding obstacles; S52, establish a safety distance value S between the drone body and surrounding obstacles, where the unit of the safety distance value S is meter. The size of the safety distance value S needs to be set in consideration of the size of the drone and the complexity of the construction inspection site; the larger the size of the drone, the larger the safety distance value S, and the more complex the construction inspection site is, the larger the safety distance value S; S53, determine the distance value e between the drone body and surrounding obstacles vt The numerical value of the safety distance value S; when e vt =S, at this time, the UAV flies according to the UAV inspection operation path data set C and maintains a safe distance value from the obstacles on the sensing side of the body; when e vt ﹤S, at this time, the UAV first flies in the opposite direction of the body sensing side to increase the distance between it and the obstacle until it is equal to the safety distance value S, and then continues to fly according to the UAV inspection operation path data set C; when e vt ﹥S, at this time the UAV continues to fly according to the UAV inspection operation path data set C.

7. A method for engineering construction inspection according to claim 6, characterized in that: The S6 comprises the following steps: S61, the drone flies to the construction inspection location according to the construction inspection site spatial grid three-dimensional coordinate data corresponding to the drone inspection operation path data cj in the drone inspection operation path data set C, and then uses the onboard camera to shoot the inspection location to generate construction inspection site feature image data 8. A method for engineering construction inspection according to claim 7, characterized in that: The S7 comprises the following steps: S71. Obtaining construction inspection site feature image data at the inspection location S72. Establish a construction inspection site quality problem image data matrix G = (g1, ..., gw, ..., gq), w = 1, 2, 3, ..., q; where g w represents the wth type of construction inspection site quality problem image data, and q represents the maximum number of types of construction inspection site quality problems; S73, using bee colony algorithm to transform the construction inspection site feature image data The construction inspection site quality problem image data matrix G contains the construction inspection site quality problem image data g w Match and identify quality problem points at the construction inspection site; The bee colony algorithm is used to transform the characteristic image data of the construction inspection site Image data of quality problems on construction inspection site w The steps for identification and matching are as follows: S731, initialize each honey source i, that is, the construction inspection site quality problem image data matrix G described in the construction inspection site quality problem image data g w Set as honey source; set the number of honey sources to η, the current number of iterations t and the maximum number of iterations T; the initial position of honey source i is w i ; S732, is the location of the honey source w i Assign hired bees. At the beginning of the search, hired bees search the construction inspection site quality problem image data matrix G in the search space by all the construction inspection site quality problem image data g w The location of the nectar source w i Around, according to the formula, search for the new nectar source location w' i , that is, traverse and search for all the construction inspection site quality problem image data g in the construction inspection site quality problem image data matrix G search space w The construction inspection site feature image data Perform image feature matching; S733, based on the new nectar source position w′ i The fitness of the nectar source is determined according to the greedy selection method, that is, the nectar source that is retained in the search space of the construction inspection site quality problem image data matrix G is searched for the image data that is consistent with the construction inspection site feature image data. Matching construction inspection site quality problem image data g w ; S734, employed bees share nectar source information, and follower bees search for the construction inspection site quality problem image data matrix G in the search space according to the probability value. Matching construction inspection site quality problem image data g w The bees follow the honey source and search for it based on the characteristic image data of the construction inspection site shared by the employed bees. Matching construction inspection site quality problem image data g w The nectar source information is used to calculate the probability of the nectar source found by the employed bees being followed according to the probability formula; S735: The follower bee searches in the same way as the employed bee and determines the nectar source i to be retained according to the greedy selection method, that is, searching for the nectar source i that matches the construction inspection site feature image data in the search space of the construction inspection site quality problem image data matrix G. Matching construction inspection site quality problem image data g w ; S736: During the random search of the construction inspection site quality problem image data matrix G in the search space, if the nectar source location w i After t iterations, the search reaches the iteration threshold No better nectar source was found. The location of the nectar source is w i The bees will be abandoned, and the corresponding employed bees will become scout bees. The scout bees will randomly generate new nectar source locations in the search space. S737: Determine whether the algorithm meets the maximum search iteration number T condition. If so, terminate and output the construction inspection site feature image data. The construction inspection site quality problem image data matrix G contains the construction inspection site quality problem image data g w If the result is a match, go to step S732; S74, when the construction inspection site feature image data Image data of quality problems on construction inspection site w If the match is successful, it means that there are quality problems in the construction inspection location corresponding to the UAV inspection operation path data cj. The UAV inspection operation path data cj and the construction inspection site feature image data of the inspection location are recorded. When the construction inspection site feature image data Image data of quality problems on construction inspection site w If the match is not successful, it means that there is no quality problem point in the construction inspection position corresponding to the UAV inspection operation path data cj, and the UAV continues to fly and perform the inspection operation according to the UAV inspection operation path data set C until the inspection operation is completed.

9. A system for implementing the method for engineering construction inspection according to any one of claims 1 to 8, characterized in that: The system includes an engineering construction inspection on-site data collection module, an unmanned aerial vehicle inspection path planning module, an unmanned aerial vehicle inspection obstacle avoidance management module, and an engineering construction inspection quality problem point data collection module.

Images