Intelligent Detection and Cleaning Method for Efflorescence Damage of Clear Water Brick Walls Based on Tethered UAVs

Abstract

The present invention discloses an intelligent detection and cleaning method for efflorescence damage of fair-faced brick walls based on a tethered unmanned aerial vehicle, which includes the steps of deploying and carrying the tethered unmanned aerial vehicle, planning the inspection flight route for efflorescence detection, collecting current waypoint data, judging whether there is efflorescence on the fair-faced brick wall, calculating the efflorescence area and judging the degree of efflorescence damage, dynamically and intelligently cleaning, and flying to the next waypoint. The present invention is suitable for the intelligent detection and cleaning of efflorescence on fair-faced brick walls in complex wall environments, realizes the precise maintenance of efflorescence damage of fair-faced brick walls, avoids the problems of incomplete cleaning or brick surface peeling and joint mortar loss caused by the traditional constant water pressure cleaning method; it also avoids the problem of water resource waste caused by constant water pressure cleaning, and has the advantages of high detection efficiency and good cleaning effect.

Description

Technical Field

[0001] The present invention relates to the technical field of construction engineering, and particularly relates to an intelligent detection and cleaning method for efflorescence damage of fair - faced brick walls based on a tethered unmanned aerial vehicle (UAV). Background Art

[0002] As an important exterior facade form of historical and modern buildings, fair - faced brick walls are widely used for their simple texture, unique masonry techniques and historical and cultural values. Affected by acid rain erosion, water vapor penetration and atmospheric pollutants, such walls are prone to efflorescence diseases, that is, white calcium hydroxide crystals precipitate on the wall surface, which not only damages the building appearance, but also causes the bricks to become loose and the bonding force of mortar to decrease, seriously threatening the structural safety. Therefore, precise cleaning and maintenance of different types of fair - faced brick walls have become a key requirement in the field of historical building protection and renovation. The current efflorescence cleaning of fair - faced brick walls mainly faces the following technical bottlenecks:

[0003] (1) High risk and low efficiency in high - altitude operations: For fair - faced brick walls of high - rise buildings or complex facades, traditional operations rely on the erection of scaffolding or high - altitude hanging baskets. Limited by the protruding shapes such as wall reliefs and moldings, it is difficult and time - consuming to install scaffolding and hanging baskets, and there are problems such as the hanging basket being unable to lift and run, as well as potential safety hazards in high - altitude operations.

[0004] (2) Coarse cleaning process, prone to secondary damage: Existing cleaning equipment (such as high - pressure water guns) lacks precise identification of the efflorescence degree. Using a unified water pressure or chemical agent for full - coverage flushing results in problems such as brick surface peeling and pointing mortar loss in lightly damaged areas due to excessive scouring, and hidden dangers remaining in severely damaged areas due to incomplete cleaning.

[0005] (3) Disconnection between detection and cleaning links, low degree of intelligence: The existing process requires manual pre - detection of efflorescence areas and then setting cleaning parameters based on experience, lacking scientific decision - making and control, and insufficient pertinence in cleaning operations. Summary of the Invention

[0006] The purpose of the present invention is to provide an intelligent detection and cleaning method for efflorescence damage of fair - faced brick walls based on a tethered UAV, so as to solve the problems of low disassembly and assembly efficiency when cleaning fair - faced brick walls on the building exterior facade by erecting scaffolding, and the inability of hanging basket equipment to lift and run due to the influence of protruding shapes such as wall reliefs and moldings, resulting in incomplete cleaning of fair - faced brick walls; to solve the problems of incomplete cleaning or brick surface peeling and pointing mortar loss in the existing cleaning methods; and to solve the problems of disconnection between the detection and cleaning links of efflorescence areas of fair - faced brick walls, resulting in insufficient pertinence and poor cleaning effect in cleaning operations.

[0007] To solve the above - mentioned technical problems, the present invention provides an intelligent detection and cleaning method for efflorescence damage of fair - faced brick walls based on a tethered UAV, including:

[0008] Step S1: Use a tethered drone with a built-in camera, ranging sensor, and AI chip, and externally equipped with a tethered cleaning module. Deploy the drone aerial triangulation calculation and close-range inspection route planning algorithm and the fair-faced brick wall efflorescence damage identification algorithm in the AI chip.

[0009] Step S2: Use the camera of the tethered drone to collect aerial photos of the building roof and facade by means of oblique photogrammetry. Solve and generate a rough building model through the aerial triangulation calculation and close-range inspection route planning algorithm deployed in the AI chip, and automatically plan a close-range inspection route.

[0010] Step S3: In the inspection route planning of the fair-faced brick wall on the building facade, when the tethered drone hovers at each waypoint, use the camera to collect images of the fair-faced brick wall on the building facade and use the ranging sensor to measure the distance data between the lens and the wall.

[0011] Step S4: The AI chip of the tethered drone uses the fair-faced brick wall efflorescence damage identification algorithm to identify and determine whether there is efflorescence on the collected image data of the fair-faced brick wall on the building facade. When there is no efflorescence, the AI chip of the tethered drone controls the externally equipped tethered cleaning module not to perform cleaning operations, and the tethered drone flies to the next waypoint. When there is efflorescence, execute Step S5.

[0012] Step S5: Calculate the efflorescence area by converting the number of pixel points of the damage target detection frame for the image with efflorescence, and judge the degree of efflorescence damage according to the efflorescence area.

[0013] Step S6: The AI chip of the tethered drone dynamically controls the water pressure, nozzle angle, and water spraying duration of the externally equipped tethered cleaning module according to the degree of efflorescence damage of the fair-faced brick wall on the building facade, and performs dynamic intelligent cleaning operations on fair-faced brick walls with different degrees of efflorescence damage.

[0014] Step S7: The tethered drone flies to the next waypoint.

[0015] Furthermore, for the intelligent detection and cleaning method of fair-faced brick wall efflorescence damage based on a tethered drone provided by the present invention, in Step S5, the method of calculating the efflorescence area by converting the number of pixel points of the damage target detection frame for the image with efflorescence includes:

[0016] According to the lens parameters of the camera of the tethered drone and the distance between the lens and the wall, use the following formula to calculate the efflorescence area by converting the number of pixel points of the damage target detection frame:

[0017]

[0018] where, b is the efflorescence areal is the distance between the lens and the wall, f is the focal length, P c is the number of pixels corresponding to 1 cm of the camera's photosensitive element, n is the number of pixel points of the damage target detection frame.

[0019] Furthermore, for the intelligent detection and cleaning method of efflorescence damage on fair - faced brick walls based on a tethered drone provided by the present invention, in step S5,

[0020] When b < 100 cm², it is determined that the degree of efflorescence damage is slight;

[0021] When 100 cm² ≤ b < 500 cm², it is determined that the degree of efflorescence damage is medium;

[0022] When b ≥ 500 cm², it is determined that the degree of efflorescence damage is severe;

[0023] where b is the efflorescence area.

[0024] Furthermore, for the intelligent detection and cleaning method of efflorescence damage on fair - faced brick walls based on a tethered drone provided by the present invention, in step S6,

[0025] When the degree of damage is slight, the AI chip of the tethered drone controls the water pressure of the externally mounted tethered cleaning module to be 5 MPa, the nozzle angle to be 15°, and the water spraying duration to be 5 s, and performs a cleaning operation on the efflorescence area of the fair - faced brick wall at the current waypoint location;

[0026] When the degree of damage is medium, the AI chip of the tethered drone controls the water pressure of the externally mounted tethered cleaning module to be 10 MPa, the nozzle angle to be 25°, and the water spraying duration to be 10 s, and performs a cleaning operation on the efflorescence area of the fair - faced brick wall at the current waypoint location;

[0027] When the degree of damage is severe, the AI chip of the tethered drone controls the water pressure of the externally mounted tethered cleaning module to be 15 MPa, the nozzle angle to be 30°, and the water spraying duration to be 15 s, and performs a cleaning operation on the efflorescence area of the fair - faced brick wall at the current waypoint location.

[0028] Furthermore, for the intelligent detection and cleaning method of efflorescence damage on fair - faced brick walls based on a tethered drone provided by the present invention, in step S2, the method for automatically planning a close - to - building inspection route includes: using the A* algorithm combined with the analysis of the curvature of the building's exterior wall to generate a zigzag - shaped close - to - building inspection route, and completing the inspection route planning for covering the fair - faced brick wall on the building's exterior wall.

[0029] Furthermore, in the intelligent detection and cleaning method for efflorescence damage of fair - faced brick walls based on a tethered drone provided by the present invention, in step S1, an efflorescence damage recognition algorithm for fair - faced brick walls is deployed in the AI chip as an object detection model trained based on a large amount of efflorescence data of fair - faced brick walls.

[0030] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0031] In the intelligent detection and cleaning method for efflorescence damage of fair - faced brick walls based on a tethered drone provided by the present invention, by using a tethered drone equipped with a tethered cleaning module, the endurance limit of traditional drones is broken through, and continuous hovering in the air is realized to intelligently detect and clean the efflorescence of fair - faced brick walls on the building facade, so as to adapt to the intelligent detection of efflorescence of fair - faced brick walls in complex wall environments with reliefs, moldings, etc. There is no need to build a scaffolding or hang a hanging basket, overcoming the problems of low disassembly and assembly efficiency of the scaffolding and the safety hazard of high - altitude falling when workers stand on the scaffolding or hanging basket for high - altitude manual detection; avoiding the problem that the hanging basket is affected by non - vertical surface shapes such as wall reliefs and moldings and cannot be lifted or lowered for manual detection and manual cleaning of fair - faced brick walls.

[0032] In the intelligent detection and cleaning method for efflorescence damage of fair - faced brick walls based on a tethered drone provided by the present invention, the AI chip built into the tethered drone automatically plans a flight path close to the inspection route, and identifies and judges whether there is efflorescence on the fair - faced brick wall by analyzing the image data of the fair - faced brick wall on the building facade collected. For the image with efflorescence, the efflorescence area is obtained by converting the number of pixel points in the damage target detection frame, and the degree of efflorescence damage is judged according to the efflorescence area. According to the degree of efflorescence damage, the water pressure, spray width and duration of the tethered cleaning module are controlled to perform dynamic cleaning operations on the efflorescence area, realizing precise maintenance of efflorescence damage of fair - faced brick walls, and avoiding the problems of incomplete cleaning or brick surface peeling and mortar loss in the joint by the traditional constant - water - pressure cleaning method; also avoiding the problem of water resource waste caused by constant - water - pressure cleaning.

[0033] In the intelligent detection and cleaning method for efflorescence damage of fair - faced brick walls based on a tethered drone provided by the present invention, the efflorescence detection and cleaning links of the fair - faced brick wall are associated, so as to perform detection and cleaning simultaneously, with the advantages of strong pertinence, high detection efficiency and good cleaning effect, avoiding the problem of poor cleaning effect caused by the inability to find the accurate efflorescence position after detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 is a flowchart of the intelligent detection and cleaning method for efflorescence damage of fair - faced brick walls based on a tethered drone. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0035] The present invention will be described in detail below with reference to the accompanying drawings: According to the following description, the advantages and features of the present invention will be clearer. It should be noted that the drawings are all in a very simplified form and use non-precise scales, only for the purpose of conveniently and clearly assisting in explaining the embodiments of the present invention.

[0036] Please refer to Figure 1 , an intelligent detection and cleaning method for efflorescence damage of fair-faced brick walls based on a tethered drone provided by an embodiment of the present invention includes:

[0037] Step S1, deployment and carrying of the tethered drone. Specifically: A tethered drone with a built-in camera, ranging sensor, and AI chip, and externally equipped with a tethered cleaning module is used. An aerial triangulation calculation and close-range inspection route planning algorithm are deployed in the AI chip, which can generate a close-range inspection route based on the rough aerial survey data of the building and deploy an efflorescence damage recognition algorithm for fair-faced brick walls, which can process images in real time to obtain the coordinates and area of the damaged area, and output a cleaning control instruction after judging the severity of efflorescence. Among them, the efflorescence damage recognition algorithm for fair-faced brick walls deployed in the AI chip is an object detection model trained based on a large amount of efflorescence data of fair-faced brick walls. The tethered cleaning module includes a modular integrated water pump (pressure adjustable), a fan-shaped nozzle array (angle ±30° electrically actuated push rod adjustment), a flow meter, and a solenoid valve, which can be connected to the chassis of the tethered drone through a quick-release interface and support a water pressure output of 5 - 15 MPa. The tethered drone is a multi-rotor tethered drone (supporting cable power supply and data transmission), equipped with a high-precision positioning module, a six-axis inertial navigation system, and an ultrasonic obstacle avoidance sensor to ensure stable hovering and route planning in a complex wall environment. The power supply and communication module of the tethered drone: 220V AC input can be realized through a high-voltage-resistant cable, and 5G / Industrial WiFi dual-mode communication is integrated to ensure low-latency transmission of control signals, image data, and cleaning parameters.

[0038] Step S2, planning of the efflorescence detection inspection route. The aerial photos of the building roof and facade are collected by the camera of the tethered drone using the oblique photogrammetry method, and the building rough model is solved and generated, and the close-range inspection route is automatically planned through the aerial triangulation calculation and close-range inspection route planning algorithm deployed in the AI chip. The method of automatically planning the close-range inspection route can adopt the A* algorithm combined with the curvature analysis of the building facade to generate a "zigzag" close-range inspection route to complete the inspection route planning covering the fair-faced brick walls of the building facade. The A* algorithm is a path planning algorithm based on heuristic search.

[0039] Step S3, current waypoint data acquisition. In the inspection route planning of the fair-faced brick wall on the exterior of a building, the tethered drone hovers at each waypoint position and collects images of the fair-faced brick wall on the exterior of the building through a camera and measures the distance data between the lens and the wall through a ranging sensor.

[0040] Step S4, determination of whether there is efflorescence on the fair-faced brick wall. The AI chip of the tethered drone uses the efflorescence damage recognition algorithm for fair-faced brick walls to identify and determine whether there is efflorescence on the collected image data of the fair-faced brick wall on the exterior of the building; when there is no efflorescence, the AI chip of the tethered drone controls the externally mounted tethered cleaning module not to perform cleaning operations, and the tethered drone flies to the next waypoint; when there is efflorescence, step S5 is executed.

[0041] Step S5, calculate the efflorescence area and determine the degree of efflorescence damage. For the image with efflorescence, the efflorescence area is obtained by converting the number of pixel points in the damage target detection box, and the degree of efflorescence damage is determined based on the efflorescence area. In step S5, in order to calculate the efflorescence area, the method of obtaining the efflorescence area by converting the number of pixel points in the damage target detection box for the image with efflorescence may include:

[0042] According to the lens parameters of the camera of the tethered drone and the distance between the lens and the wall, the following formula is used to convert the number of pixel points in the damage target detection box to obtain the efflorescence area:

[0043]

[0044] where, b is the efflorescence area, l is the distance between the lens and the wall, f is the focal length, P c is the number of pixels corresponding to 1 cm of the camera's photosensitive element, n is the number of pixel points in the damage target detection box. Among them f is the focal length, P c The number of pixels corresponding to 1 cm of the camera's photosensitive element is the lens parameter.

[0045] In step S5, in order to determine the degree of damage of the efflorescence area.

[0046] When b < 100 cm², the degree of efflorescence damage is slight.

[0047] When 100 cm² ≤ b < 500 cm², the degree of efflorescence damage is medium.

[0048] When b ≥ 500 cm², the degree of efflorescence damage is severe.

[0049] Step S6, dynamic intelligent cleaning. The AI chip of the tethered drone dynamically controls the water pressure, nozzle angle, and water spraying duration of the externally mounted tethered cleaning module according to the degree of efflorescence damage of the fair-faced brick wall on the building facade, and performs dynamic intelligent cleaning operations on fair-faced brick walls with different degrees of efflorescence damage.

[0050] In step S6, in order to intelligently clean the efflorescence area of the fair-faced brick wall.

[0051] When the degree of damage is slight, the AI chip of the tethered drone controls the water pressure of the externally mounted tethered cleaning module to be 5 MPa, the nozzle angle to be 15°, and the water spraying duration to be 5 s, and performs a cleaning operation on the efflorescence area of the fair-faced brick wall at the current waypoint location.

[0052] When the degree of damage is medium, the AI chip of the tethered drone controls the water pressure of the externally mounted tethered cleaning module to be 10 MPa, the nozzle angle to be 25°, and the water spraying duration to be 10 s, and performs a cleaning operation on the efflorescence area of the fair-faced brick wall at the current waypoint location.

[0053] When the degree of damage is severe, the AI chip of the tethered drone controls the water pressure of the externally mounted tethered cleaning module to be 15 MPa, the nozzle angle to be 30°, and the water spraying duration to be 15 s, and performs a cleaning operation on the efflorescence area of the fair-faced brick wall at the current waypoint location.

[0054] Step S7, the tethered drone flies to the next waypoint.

[0055] The intelligent detection and cleaning method for efflorescence damage of fair-faced brick walls based on tethered drones provided by the embodiments of the present invention uses a tethered cleaning module carried by a tethered drone to break through the endurance limit of traditional drones, realizes continuous hovering in the air for intelligent detection and intelligent cleaning of the efflorescence condition of fair-faced brick walls on the building facade, so as to adapt to the intelligent detection of efflorescence of fair-faced brick walls in complex wall environments such as reliefs and moldings. There is no need to build a scaffolding or hang a hanging basket, which overcomes the low disassembly and assembly efficiency of the scaffolding and the safety hazard of high-altitude falling when workers stand on the scaffolding or hanging basket for high-altitude manual detection; it avoids the problem that the hanging basket is affected by non-vertical surface shapes such as wall reliefs and moldings and cannot be lifted and lowered to perform manual detection and manual cleaning of fair-faced brick walls.

[0056] The intelligent detection and cleaning method for efflorescence damage of fair - faced brick walls based on a tethered drone provided by an embodiment of the present invention automatically plans a close - range inspection route through the AI chip built into the tethered drone, and identifies the collected image data of the fair - faced brick walls on the building facade to determine whether there is efflorescence on the fair - faced brick walls. For the images with efflorescence, the efflorescence area is obtained by converting the number of pixel points in the damage target detection frame. The degree of efflorescence damage is judged according to the efflorescence area, and the water pressure, spray width and duration of the tethered cleaning module are controlled according to the degree of efflorescence damage to perform dynamic cleaning operations on the efflorescence area, realizing the precise maintenance of the efflorescence damage of fair - faced brick walls, and avoiding the problems of incomplete cleaning or brick surface peeling and mortar loss in the joint by the traditional constant - water - pressure cleaning method; it also avoids the problem of water resource waste caused by constant - water - pressure cleaning.

[0057] The intelligent detection and cleaning method for efflorescence damage of fair - faced brick walls based on a tethered drone provided by an embodiment of the present invention associates the efflorescence detection and cleaning links of fair - faced brick walls, so as to perform detection while cleaning, with the effects of sufficient pertinence, high detection efficiency and good cleaning effect, and avoid the problem of poor cleaning effect caused by the inability to find the accurate efflorescence position after detection.

[0058] The intelligent detection and cleaning method for efflorescence damage of fair - faced brick walls based on a tethered drone provided by an embodiment of the present invention realizes the integrated operation of wall cruising, efflorescence detection, damage grading and precise cleaning, providing an intelligent solution for the efflorescence cleaning of fair - faced brick walls in historical buildings, so as to solve the core problems of "difficult to identify, rough control and low efficiency" in the efflorescence cleaning of fair - faced brick walls.

[0059] The present invention is not limited to the above - mentioned specific embodiments. Obviously, the above - described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. Based on the described embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art belong to the scope of protection of the present invention. Those skilled in the art can make other levels of modifications and changes to the present invention. Thus, if these modifications and changes of the present invention are within the scope of the claims of the present invention, the present invention also intends to include these modifications and changes.

Claims

1. An intelligent detection and cleaning method for efflorescence damage of fair - faced brick walls based on tethered drones, characterized in that, Including: Step S1: Use a tethered drone with a built-in camera, ranging sensor, and AI chip, and a tethered cleaning module externally mounted. Deploy the drone aerial triangulation calculation and close-range inspection route planning algorithm and the fair-faced brick efflorescence damage identification algorithm in the AI chip. Step S2: Use the camera of the tethered drone to collect aerial photos of the building roof and facade by means of oblique photogrammetry. Solve and generate a rough building model and automatically plan a close-range inspection route through the aerial triangulation calculation and close-range inspection route planning algorithm deployed in the AI chip. Step S3: In the inspection route planning of the fair-faced brick wall on the building facade, the tethered drone hovers at each waypoint position, collects images of the fair-faced brick wall on the building facade through the camera, and measures the distance data between the lens and the wall through the ranging sensor. Step S4: The AI chip of the tethered drone uses the fair-faced brick efflorescence damage identification algorithm to identify and determine whether there is efflorescence on the collected image data of the fair-faced brick wall on the facade. When there is no efflorescence, the AI chip of the tethered drone controls the externally mounted tethered cleaning module not to perform the cleaning operation, and the tethered drone flies to the next waypoint. When there is efflorescence, perform Step S5. Step S5: Calculate the efflorescence area by converting the number of pixel points in the damage target detection frame for the image with efflorescence, and judge the degree of efflorescence damage according to the efflorescence area. Step S6: The AI chip of the tethered drone dynamically controls the water pressure, nozzle angle, and spraying duration of the externally mounted tethered cleaning module according to the degree of efflorescence damage of the fair-faced brick wall on the building facade to perform dynamic intelligent cleaning operations on fair-faced brick walls with different degrees of efflorescence damage. Step S7: The tethered drone flies to the next waypoint.

2. The intelligent detection and cleaning method for efflorescence damage of fair - faced brick walls based on tethered drones according to claim 1, wherein, In Step S5, the method of calculating the efflorescence area by converting the number of pixel points in the damage target detection frame for the image with efflorescence includes: According to the lens parameters of the camera of the tethered drone and the distance between the lens and the wall, use the following formula to calculate the efflorescence area by converting the number of pixel points in the damage target detection frame: ； Among them, b is the area of efflorescence, l is the distance between the lens and the wall, and f is the focal length, P c is the number of pixels corresponding to 1 cm of the camera's photosensitive element, n is the number of pixel points of the damage target detection frame.

3. The intelligent detection and cleaning method for efflorescence damage of fair-faced brick walls based on a tethered unmanned aerial vehicle according to claim 1, wherein, In Step S5, When b < 100 cm², judge that the degree of efflorescence damage is slight; When 100 cm² ≤ b < 500 cm², judge that the degree of efflorescence damage is medium; When b ≥ 500 cm², judge that the degree of efflorescence damage is severe; where b is the efflorescence area.

4. The intelligent detection and cleaning method for efflorescence damage of fair - faced brick walls based on tethered drones according to claim 3, characterized in that, In Step S6, When the degree of damage is slight, the AI chip of the tethered drone controls the water pressure of the externally mounted tethered cleaning module to be 5 MPa, the nozzle angle to be 15°, and the spraying duration to be 5 s, and performs cleaning operations on the efflorescence area of the fair-faced brick wall at the current waypoint position. When the degree of damage is medium, the AI chip of the tethered drone controls the water pressure of the externally mounted tethered cleaning module to be 10 MPa, the nozzle angle to be 25°, and the spraying duration to be 10 s, and performs cleaning operations on the efflorescence area of the fair-faced brick wall at the current waypoint position. When the degree of damage is severe, the AI chip of the tethered drone controls the water pressure of the externally mounted tethered cleaning module to be 15 MPa, the nozzle angle to be 30°, and the water spraying duration to be 15 s, and performs a cleaning operation on the area of the efflorescence on the fair-faced brick wall at the current waypoint.

5. The intelligent detection and cleaning method for efflorescence damage of fair-faced brick walls based on tethered drones according to claim 1, wherein, In step S2, the method for automatically planning a close patrol route includes: using the A* algorithm combined with the analysis of the curvature of the building facade to generate a "zigzag" close patrol route, and completing the patrol route planning for covering the fair-faced brick wall of the building facade.

6. The intelligent detection and cleaning method for efflorescence damage of fair-faced brick walls based on tethered drones according to claim 1, wherein, In step S1, a fair-faced brick wall efflorescence damage recognition algorithm is deployed in the AI chip as an object detection model trained based on a large amount of fair-faced brick wall efflorescence data.

Images