Body pit detection device and detection method based on point cloud data

Abstract

The application relates to a body pit detection device and method based on point cloud data, which comprises the following modules: a point cloud acquisition module for acquiring body surface point cloud data; a point cloud processing module for carrying out point cloud processing on the body surface point cloud data; a point cloud reconstruction module for carrying out point cloud reconstruction to obtain a three-dimensional CAD model of the body surface; a target detection module for extracting body surface pit data to obtain pit information; a pit discrimination module for determining pit repairability based on the pit information to generate a body surface pit detection result and body repair position information; a result output module for outputting and displaying information; and a remote control module for information interaction with the point cloud acquisition module through a wireless communication mode. The application can be used for detecting and recording pit damage after impact on an aircraft body surface, can reduce the visual inspection workload of aircraft maintenance personnel, and can effectively improve detection accuracy.

Description

Technical Field

[0001] This invention relates to the field of intelligent inspection technology for aircraft airframe structures, specifically to an airframe dent detection device and method based on point cloud data. Background Technology

[0002] During takeoff, flight, landing, and maintenance, aircraft skin, panels, and fairings may develop permanent or localized dents due to foreign object impacts, overloading, or improper maintenance, thus forming fuselage dents. Fuse dents increase aircraft drag and affect flight performance. Furthermore, under loading conditions, the secondary bending stress generated at fuselage dents is significant (similar to stress concentration), leading to decreased structural fatigue strength and impacting aircraft safety. Therefore, the detection and tracking of fuselage dents is a crucial aspect of ground maintenance.

[0003] The permissible damage tolerance values ​​for fuselage dents on different components of the aircraft fuselage vary. For example, the area around the leading edge of the wing or the static pressure port is an area with strict aerodynamic requirements and cannot be judged according to the general fuselage dent damage tolerance. Instead, it needs to be judged according to the aerodynamic requirements. On the other hand, fuselage dents on components such as skin, panels, fairings, and web components are considered permissible damage, and the maximum depth of the fuselage dent must be less than a certain value, and the waviness (A / Y) of the fuselage dent must be greater than a certain value.

[0004] Body dents belong to Category 1 Barely Visually Detectable Defects (BVID), which are the smallest defects that can be reliably detected using the designed visual inspection methods. Currently, body dents are inspected using direct visual inspection, which involves examining the body directly with the human eye or with equipment such as a magnifying glass. Direct visual inspection is economical, simple, and fast. However, the inspection results are significantly affected by the subjective factors of the inspector. The shape and size of the defect, lighting conditions, testing distance, and the age of the inspector all have a direct impact on the inspection results.

[0005] Patent application number CN202111218278.2 discloses "a method, system and unmanned truck for identifying potholes on mining roads". This patent focuses on the pothole identification method, which is different from the present invention. The identification system has a large granularity and does not involve the identification device or the information judgment after the pothole is identified.

[0006] Patent application number CN202210677801.6 discloses "A method for damage detection and 2D-3D positioning of aircraft skin images". This patent uses image processing and machine learning methods for damage detection, focusing on the detection method, without involving the detection system and device, which is different from the stage of this invention.

[0007] Patent application number CN202021134899.3 discloses "An aircraft dent detection system". This patent is a utility model patent and mainly provides an aircraft dent detection system, including a lifting mechanism, a gimbal, a three-dimensional information acquisition module, a control module, a CMOS camera, an excitation source and an infrared thermal imager. This patent focuses on the dent detection system and detection device, and does not involve the detection method, process and maintenance information output. Furthermore, the modules covered by the detection system and the information transmission process are different from those of this invention.

[0008] Patent application number CN202221023976.7 discloses "a detection device and a pit flushing device." This patent is a utility model patent, mainly involving a detection device and a pit flushing device. The purpose of this patent is to improve the inspection accuracy of pits, thereby improving product quality. This solution focuses on the detection device and does not involve the detection method, process, or system, which differs from this invention.

[0009] Patent application number CN202211061350.X discloses "a quantitative analysis method and system for defects in the surface of automotive exterior panels". This patent is an invention patent and mainly relates to a defect detection method for automotive exterior panels. This patent does not consider point cloud data compensation, so its adaptability to the environment is relatively limited. In addition, this patent does not involve a detection result output module.

[0010] Patent application number 202210983494.4 discloses "a method for detecting and identifying defects in the fan-shaped section blade of an aero-engine". This patent is an invention patent and mainly involves the detection of defects in the fan-shaped section blade of an engine. After data acquisition, the patent does not consider point cloud data compensation, which has limited adaptability to the environment and results in inaccurate data. The defects are obtained by calculating geometric feature parameters through formulas, without a comparison model, which increases the amount of calculation and results in low accuracy.

[0011] Patent application number 202210681120.7 discloses "a rapid detection device for pits and cracks on the surface of large stone slabs". This patent is an invention patent and mainly relates to a defect detection device for the surface of large stone slabs. This patent uses a sliding rail robotic arm, which cannot achieve six degrees of freedom of movement and is suitable for flat parts inspection. This patent combines three-dimensional scanning and two-dimensional imaging, which reduces the detection efficiency.

[0012] The information disclosed in this background section is intended only to enhance the understanding of the general background of the invention and should not be construed as an admission or in any way implying that the information constitutes prior art known to those skilled in the art. Summary of the Invention

[0013] In view of the shortcomings of the existing technology, the purpose of this invention is to provide an airframe dent detection device and method based on point cloud data, to solve the influence of subjective factors of personnel during visual inspection, improve the accuracy of detection, and better ensure the safety of aircraft.

[0014] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0015] The body dent detection equipment based on point cloud data is characterized by comprising: a point cloud acquisition module, a point cloud processing module, a point cloud reconstruction module, a target detection module, a dent discrimination module, a result output module, and a remote control module.

[0016] The point cloud acquisition module is used to acquire point cloud data of the machine body surface;

[0017] The point cloud processing module is used to process point cloud data on the surface of the machine body.

[0018] The point cloud reconstruction module is used to reconstruct the processed point cloud data to obtain a three-dimensional CAD model of the body surface.

[0019] The target detection module is used to extract pit data from the 3D CAD model of the aircraft surface to obtain pit information;

[0020] The pit identification module is used to determine the repairability of pits based on pit information, and to generate pit detection results on the machine surface and machine repair location information.

[0021] The results output module is used to output and display the detection results of pits on the machine body surface and the machine body repair location information;

[0022] The remote control module and the point cloud acquisition module exchange information via wireless communication.

[0023] Based on the above technical solution, the point cloud acquisition module includes: a six-degree-of-freedom robotic arm, a monitoring station, a binocular camera, a decoder, and a point cloud acquisition card;

[0024] The six-degree-of-freedom robotic arm includes: three translation servo motors and matching translation servo motor drivers, used to realize translation along the three coordinate axes OX, OY, and OZ; three rotation servo motors and matching rotation servo motor drivers, used to realize rotation along the three coordinate axes OX, OY, and OZ; the robotic arm is used to guide the monitoring station to move along the detection trajectory information.

[0025] The servo motor driver is used to receive detection trajectory information of different aircraft models sent by the remote control module;

[0026] A binocular camera, a decoder, and a point cloud acquisition card are installed on the monitoring console. The binocular camera performs six degrees of freedom transformations with the monitoring console to acquire point cloud data of the aircraft surface. The decoder is used to control the binocular camera to focus and obtain clear point cloud data of the aircraft surface. The point cloud acquisition card is used to store the point cloud data of the aircraft surface and to transmit the stored point cloud data of the aircraft surface to the point cloud processing module.

[0027] Based on the above technical solution, the point cloud acquisition module also includes: a brightness sensor, an optocoupler, a relay, and structured light;

[0028] A brightness sensor is used to send the collected ambient brightness signal to a decoder, which generates a corresponding structured light control signal based on the ambient brightness signal. An optocoupler is used to receive the structured light control signal sent by the decoder and control a relay based on the structured light control signal. The relay is used to turn the structured light on or off and to adjust the brightness of the structured light. The structured light is used to illuminate the surface of the body facing the binocular camera.

[0029] Based on the above technical solution, the point cloud processing module includes: a point cloud compensation unit, a noise processing unit, an adaptive repair unit, and a simplification processing unit;

[0030] The point cloud compensation unit is used to compensate for atmospheric parameters in the point cloud data of the aircraft surface;

[0031] The noise processing unit is used to filter, denoise, and smooth the point cloud data of the machine surface after compensation processing;

[0032] The adaptive repair unit is used to perform B-spline interpolation adaptive repair on the noise-reduced point cloud data of the aircraft surface;

[0033] The simplification processing unit is used to simplify the feature values ​​of the repaired body surface point cloud data, and the simplified body surface point cloud data is transmitted to the point cloud reconstruction module.

[0034] The point cloud compensation unit specifically includes: an atmospheric temperature compensation module, an atmospheric pressure compensation module, an atmospheric humidity compensation module, and an airflow compensation module;

[0035] Each compensation module sequentially performs temperature compensation, pressure compensation, humidity compensation, and airflow compensation on the point cloud data of the machine surface.

[0036] Based on the above technical solution, the point cloud reconstruction module includes: a registration and stitching unit and a surface reconstruction unit;

[0037] The registration and stitching unit performs registration and stitching processing on the simplified point cloud data of the body surface through a coordinate transformation matrix;

[0038] The surface reconstruction unit uses the least squares method to reconstruct the surface of the aircraft surface point cloud data after registration and stitching, and obtains a three-dimensional CAD model of the aircraft surface.

[0039] Based on the above technical solution, the target detection module includes: a size unification unit, a pit detection unit, a pit data extraction unit, and a data output unit;

[0040] The dimension unification unit is used to unify the dimensions of the 3D CAD model of the aircraft surface.

[0041] The pit detection unit is used to identify and detect pits on the 3D CAD model of the body surface after the dimensions have been uniformly processed.

[0042] The pit data extraction unit is used to extract data from each pit that has been identified and detected. The data extraction includes the extraction of the following pit information: the minimum width W of the pit, the depth Y at the width W, the ratio of the width W to the depth Y (i.e., the depth ratio), the structure name corresponding to the pit, and the part number corresponding to the pit.

[0043] The data output unit is used to transmit the extracted pit information to the pit discrimination module.

[0044] Based on the above technical solution, the pit discrimination module includes: a pit classification unit, a pit determination unit, and a result loading unit;

[0045] The pit classification unit is used to classify pits into two categories based on pit information: mandatory pits and pits to be determined.

[0046] The pit determination unit is used to determine whether the pit to be determined is a pit that needs to be repaired;

[0047] The result loading unit is used to summarize the information obtained synchronously with the point cloud data and correspond the information with the dents that must be repaired and the dents that need to be repaired; then it forms complete dent information by combining the location, section, part number and corresponding material of the dents on the machine body, and transmits the complete information to the result output module.

[0048] Based on the above technical solution, the result output module includes: an LED display screen, a power button, an alarm indicator light, and an alarm button;

[0049] The LED display screen is used to display complete information about the dents;

[0050] The power button is used to control the power supply to the LED display screen;

[0051] The alarm indicator light is used to alert the user to dents that require repair or need to be repaired.

[0052] The alarm button is used to turn off the alarm indicator light and sends the pressed status to the LED display screen for display.

[0053] The fuselage dent detection system based on point cloud data is characterized by comprising: according to the division of aircraft segments, each segment is equipped with one of the aforementioned point cloud data-based fuselage dent detection devices in the following manner:

[0054] A corresponding number of sliding rails are installed on the ground according to the number of sections;

[0055] Each slide rail is equipped with a mounting frame. Two body dent detection devices based on point cloud data are installed in a set, with each set corresponding to a mounting frame.

[0056] The mounting frame drives the body pit detection equipment based on point cloud data to move back and forth along the slide rail;

[0057] One set of point cloud data-based fuselage dent detection equipment is used to detect the upper surface of aircraft sections, while the other set is used to detect the lower surface of aircraft sections.

[0058] The method for detecting body pits based on point cloud data is characterized by the following specific steps:

[0059] After the point cloud data-based body pit detection equipment described above is powered on, it acquires point cloud data of the body surface through the point cloud acquisition module;

[0060] The point cloud data of the machine surface acquired by the point cloud acquisition module is stored using a point cloud acquisition card;

[0061] The atmospheric parameters of the body surface point cloud data stored in the storage unit are compensated by the point cloud compensation unit in the point cloud processing module.

[0062] The noise processing unit in the point cloud processing module performs filtering, noise reduction, and smoothing processing on the body surface point cloud data after compensation by the point cloud compensation unit.

[0063] The adaptive repair unit in the point cloud processing module performs B-spline interpolation adaptive repair on the noise-processed and smoothed point cloud data of the body surface by the noise processing unit.

[0064] The feature value simplification process is performed on the point cloud data of the body surface repaired by the adaptive repair unit through the simplification processing unit in the point cloud processing module.

[0065] The registration and stitching unit in the point cloud reconstruction module performs registration and stitching processing on the body surface point cloud data processed by the point cloud processing module.

[0066] The surface reconstruction unit in the point cloud reconstruction module uses the least squares method to reconstruct the surface of the entire machine from the point cloud data of the registration and stitching unit.

[0067] The size unification unit in the target detection module is used to unify the size of the 3D CAD model of the body after the point cloud reconstruction.

[0068] The pit detection unit in the target detection module is used to identify and detect pits in the 3D CAD model after the dimensions are unified.

[0069] The pit data extraction unit in the target detection module extracts the data of the identified pits;

[0070] The pit data is transmitted to the pit discrimination module through the data output unit in the target detection module.

[0071] The pit data obtained is classified by the pit classification unit in the pit discrimination module.

[0072] The pit identification module uses a pit determination unit to determine the pits to be repaired and the location of the pits on the machine body that need repair, based on pit maintainability standards.

[0073] The result loading unit in the dent identification module merges the location of the dents on the machine body to be repaired into complete information.

[0074] The results obtained from the pit detection module are displayed through the result output module.

[0075] The device and method for detecting body dents based on point cloud data described in this invention have the following beneficial effects:

[0076] 1. This invention uses a binocular camera to collect data on the surface of an aircraft, obtaining point cloud data of the aircraft surface. By processing the point cloud data, the final data of the aircraft body pits is obtained, which improves the accuracy of detection and better ensures the safety of the aircraft.

[0077] 2. This invention can avoid the influence of subjective factors such as defect shape and size, lighting conditions, and test distance on the tester, which in turn affects the accuracy of the results, and can achieve more accurate and faster detection of body pits.

[0078] 3. This invention can be used for testing when the aircraft is grounded, saving manpower and time resources.

[0079] 4. This invention detects dent defects in the machine body by collecting point cloud data. It also considers point cloud data compensation and makes the collected data more accurate through data processing. The results are more precise by comparing with the original digital model. It can adapt to various working environments with different temperatures, humidity and airflow, and can detect products of different shapes.

[0080] 5. This invention uses a six-degree-of-freedom robotic arm for scanning, which can meet the needs of different aircraft models and is also applicable to other products with dent detection requirements. It compares three-dimensional point cloud data in one go, making it more efficient.

[0081] 6. This invention achieves intelligent detection and intelligent result transmission through a result output module. Intelligent detection means that the detection device described in this invention can intelligently identify locations requiring repair without requiring manual dent detection. Intelligent result transmission means that the output result includes necessary repair-related information, such as the location requiring repair, component number, and component material.

[0082] The point cloud data-based dent detection equipment and method described in this invention can be used for the detection and recording of dent damage on the surface of an aircraft after impact. It can reduce the workload of visual inspection by aircraft maintenance personnel and effectively improve the accuracy of detection. Attached Figure Description

[0083] The present invention includes the following figures:

[0084] The accompanying drawings are provided to better understand the invention and are not intended to unduly limit the scope of the invention. Wherein:

[0085] Figure 1 The system architecture diagram of the body pit detection device based on point cloud data described in this invention.

[0086] Figure 2 Schematic diagram of point cloud acquisition module.

[0087] Figure 3 Schematic diagram of point cloud processing module.

[0088] Figure 4 Schematic diagram of point cloud compensation unit.

[0089] Figure 5 Schematic diagram of point cloud reconstruction module.

[0090] Figure 6 Schematic diagram of the target detection module.

[0091] Figure 7 Diagram of the pit.

[0092] Figure 8 Schematic diagram of the pit detection module.

[0093] Figure 9 Schematic diagram of the result output module.

[0094] Figure 10 The flowchart of the body pit detection method based on point cloud data described in this invention. Detailed Implementation

[0095] The present invention will be further described in detail below with reference to the accompanying drawings. This detailed description is an illustration in conjunction with exemplary embodiments of the invention, including various details of the embodiments to aid understanding, and should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the invention. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0096] like Figure 1 As shown, the present invention provides a body dent detection device based on point cloud data, including: a point cloud acquisition module, a point cloud processing module, a point cloud reconstruction module, a target detection module, a dent discrimination module, a result output module, and a remote control module;

[0097] The point cloud acquisition module is used to acquire point cloud data of the machine body surface;

[0098] like Figure 2 As shown, the point cloud acquisition module includes: a six-degree-of-freedom robotic arm, a monitoring station, a binocular camera, a decoder, and a point cloud acquisition card;

[0099] The six-degree-of-freedom robotic arm includes: three translation servo motors and matching translation servo motor drivers, used to realize translation along the three coordinate axes OX, OY, and OZ; three rotation servo motors and matching rotation servo motor drivers, used to realize rotation along the three coordinate axes OX, OY, and OZ; the robotic arm is used to guide the monitoring station to move along the detection trajectory information.

[0100] The servo motor driver is used to receive detection trajectory information of different aircraft models sent by the remote control module. Different aircraft models (if the aircraft model information is different, they are considered as different aircraft models) correspond to different detection trajectory information.

[0101] A binocular camera, a decoder, and a point cloud acquisition card are installed on the monitoring console. The binocular camera performs six degrees of freedom transformations with the monitoring console to acquire point cloud data of the aircraft surface. The decoder is used to control the binocular camera to focus and obtain clear point cloud data of the aircraft surface. The point cloud acquisition card is used to store the point cloud data of the aircraft surface and to transmit the stored point cloud data of the aircraft surface to the point cloud processing module.

[0102] As one of the alternative implementation schemes, the point cloud acquisition module also includes: a brightness sensor, an optocoupler, a relay, and structured light;

[0103] A brightness sensor is used to send the collected ambient brightness signal to a decoder, which generates a corresponding structured light control signal based on the ambient brightness signal. An optocoupler is used to receive the structured light control signal sent by the decoder and control a relay based on the structured light control signal. The relay is used to turn the structured light on or off and to adjust the brightness of the structured light. The structured light is used to illuminate the surface of the body facing the binocular camera.

[0104] The point cloud processing module is used to process point cloud data on the surface of the machine body.

[0105] like Figure 3 As shown, the point cloud processing module includes: a point cloud compensation unit, a noise processing unit, an adaptive repair unit, and a simplification processing unit;

[0106] The point cloud compensation unit is used to compensate for atmospheric parameters in the point cloud data of the aircraft surface;

[0107] The noise processing unit is used to filter, denoise, and smooth the point cloud data of the machine surface after compensation processing;

[0108] The adaptive repair unit is used to perform B-spline interpolation adaptive repair on the noise-reduced point cloud data of the aircraft surface;

[0109] The simplification processing unit is used to simplify the feature values ​​of the repaired body surface point cloud data, and the simplified body surface point cloud data is transmitted to the point cloud reconstruction module.

[0110] As one of the alternative implementation schemes, such as Figure 4 As shown, the point cloud compensation unit specifically includes: an atmospheric temperature compensation module, an atmospheric pressure compensation module, an atmospheric humidity compensation module, and an airflow compensation module;

[0111] Each compensation module sequentially performs temperature compensation, pressure compensation, humidity compensation, and airflow compensation on the point cloud data of the machine surface;

[0112] As one of the alternative implementation schemes, the point cloud processing module further includes: a storage unit;

[0113] The storage unit is used to back up the surface point cloud data of the machine body stored in the point cloud acquisition card, and to store the data processing results of each module; the storage unit can be selected according to actual needs;

[0114] The point cloud reconstruction module is used to reconstruct the processed point cloud data to obtain a three-dimensional CAD model of the body surface.

[0115] like Figure 5 As shown, the point cloud reconstruction module includes: a registration and stitching unit and a surface reconstruction unit;

[0116] The registration and stitching unit performs registration and stitching processing on the simplified point cloud data of the body surface through a coordinate transformation matrix;

[0117] The surface reconstruction unit performs surface reconstruction on the registered and stitched point cloud data of the aircraft surface using the least squares method to obtain a three-dimensional CAD model of the aircraft surface.

[0118] The target detection module is used to extract pit data from the 3D CAD model of the aircraft surface to obtain pit information;

[0119] like Figure 6 As shown, the target detection module includes: a size unification unit, a pit detection unit, a pit data extraction unit, and a data output unit;

[0120] The dimension unification unit is used to unify the dimensions of the 3D CAD model of the aircraft surface.

[0121] The pit detection unit is used to identify and detect pits on the 3D CAD model of the body surface after the dimensions have been uniformly processed.

[0122] The pit data extraction unit is used to extract data from each identified and detected pit. The data extraction includes extracting the following pit information: the minimum width W of the pit, the depth Y at width W, the ratio of width W to depth Y (depth ratio), the structure name corresponding to the pit, and the part number corresponding to the pit. A schematic diagram of the pit is shown below. Figure 7 As shown; the information extracted above originates from information obtained synchronously with point cloud data;

[0123] The data output unit is used to transmit the extracted pit information to the pit discrimination module;

[0124] The pit identification module is used to determine the repairability of pits based on pit information, and to generate pit detection results on the machine surface and machine repair location information.

[0125] like Figure 8 As shown, the pit discrimination module includes: a pit classification unit, a pit determination unit, and a result loading unit;

[0126] The pit classification unit is used to classify pits into two categories based on pit information: mandatory pits and pits to be determined. For example, pits located in the most critical aerodynamic influence zone are mandatory pits. The most critical aerodynamic influence zone includes, but is not limited to, the atmospheric data sensor and its vicinity.

[0127] The dent identification unit is used to determine whether a dent to be identified is one that needs repair. For example, it determines whether a dent to be identified needs repair based on dent repair standards. The dent repair standards are: collecting the depth Y at a width of W and calculating the depth ratio W / Y. For example, the maximum allowable dent Y value for the fuselage is 3.175mm (0.125 inches), and the depth ratio must be greater than or equal to 30. Based on the above standards, it determines whether a dent to be identified needs repair, and obtains the location of the fuselage dent that needs repair based on the determination result.

[0128] The result loading unit is used to summarize the information obtained synchronously with the point cloud data and correspond the information with the required dents and the dents that need to be repaired; then it forms complete dent information by combining the location, section, part number and corresponding material of the dents on the machine body, and transmits the complete information to the result output module.

[0129] The results output module is used to output and display the detection results of pits on the machine body surface and the machine body repair location information;

[0130] like Figure 9 As shown, the result output module includes: an LED display screen, a power button, an alarm indicator light, and an alarm button;

[0131] The LED display screen is used to display complete information about the dents;

[0132] The power button is used to control the power supply to the LED display screen;

[0133] The alarm indicator light is used to alert the user to dents that require repair or need to be repaired.

[0134] The alarm button is used to turn off the alarm indicator light and sends the pressed status to the LED display screen for display.

[0135] The remote control module and the point cloud acquisition module interact with each other via wireless communication. The information interaction includes: sending instruction information to the point cloud acquisition module and receiving real-time point cloud acquisition information sent by the point cloud acquisition module.

[0136] The remote control module sends command information to the translation servo motor driver and rotation servo motor driver in the point cloud acquisition module.

[0137] Based on the above technical solutions, the present invention further provides a fuselage dent detection system based on point cloud data, comprising: according to the division of aircraft segments, each segment is equipped with one of the above-mentioned fuselage dent detection devices based on point cloud data in the following manner: for example, the aircraft segments are divided into the nose, forward fuselage, mid-fuselage, wing, mid-rear fuselage, rear fuselage, vertical tail, horizontal tail, etc.; each segment corresponds to its own detection trajectory information;

[0138] A corresponding number of sliding rails are installed on the ground according to the number of sections;

[0139] Each slide rail is equipped with a mounting frame. Two body dent detection devices based on point cloud data are installed in a set, with each set corresponding to a mounting frame.

[0140] The mounting frame drives the body pit detection equipment based on point cloud data to move back and forth along the slide rail;

[0141] One set of point cloud data-based fuselage dent detection equipment is used to detect the upper surface of aircraft sections, while the other set is used to detect the lower surface of aircraft sections.

[0142] This allows for comprehensive inspection without blind spots and significantly improves inspection efficiency; it can also serve as a backup for each other, extending equipment lifespan and reducing the impact of malfunctions on inspection progress.

[0143] like Figure 10 As shown, this invention provides a method for detecting body pits based on point cloud data, and the specific steps are as follows:

[0144] After the above-mentioned point cloud data-based aircraft pit detection equipment is powered on, it acquires point cloud data of the aircraft surface through the point cloud acquisition module; by default, it detects all outer surface skins of the aircraft.

[0145] The point cloud data of the machine surface acquired by the point cloud acquisition module is stored using a point cloud acquisition card;

[0146] The atmospheric parameters of the body surface point cloud data stored in the storage unit are compensated by the point cloud compensation unit in the point cloud processing module.

[0147] The noise processing unit in the point cloud processing module performs filtering, noise reduction, and smoothing processing on the body surface point cloud data after compensation by the point cloud compensation unit.

[0148] The adaptive repair unit in the point cloud processing module performs B-spline interpolation adaptive repair on the noise-processed and smoothed point cloud data of the body surface by the noise processing unit.

[0149] The feature value simplification process is performed on the point cloud data of the body surface repaired by the adaptive repair unit through the simplification processing unit in the point cloud processing module.

[0150] The registration and stitching unit in the point cloud reconstruction module performs registration and stitching processing on the body surface point cloud data processed by the point cloud processing module.

[0151] The surface reconstruction unit in the point cloud reconstruction module uses the least squares method to reconstruct the surface of the entire machine from the point cloud data of the registration and stitching unit.

[0152] The size unification unit in the target detection module is used to unify the size of the 3D CAD model of the body after the point cloud reconstruction.

[0153] The pit detection unit in the target detection module is used to identify and detect pits in the 3D CAD model after the dimensions are unified.

[0154] The pit data extraction unit in the target detection module extracts the data of the identified pits;

[0155] The pit data is transmitted to the pit discrimination module through the data output unit in the target detection module.

[0156] The pit data obtained is classified by the pit classification unit in the pit discrimination module.

[0157] The pit identification module uses a pit determination unit to determine the pits to be repaired and the location of the pits on the machine body that need repair, based on pit maintainability standards.

[0158] The result loading unit in the dent identification module merges the location of the dents on the machine body to be repaired into complete information.

[0159] The results obtained from the pit detection module are displayed through the result output module.

[0160] The specific algorithm can be a well-known one, and will not be described in detail here.

[0161] As one of the alternative implementation schemes, after each machine body pit detection device based on point cloud data completes the detection, the point cloud data is stitched together to form a three-dimensional CAD model of the entire machine surface.

[0162] The contents not described in detail in this specification are existing technologies known to those skilled in the art.

[0163] The above description is only a preferred embodiment of the present invention. The scope of protection of the present invention is not limited to the above embodiments. Any equivalent modifications or changes made by those skilled in the art based on the content disclosed in the present invention should be included in the scope of protection set forth in the claims.

Claims

1. A device for detecting body dents based on point cloud data, characterized in that, include: The system includes a point cloud acquisition module, a point cloud processing module, a point cloud reconstruction module, a target detection module, a pit discrimination module, a result output module, and a remote control module. The point cloud acquisition module is used to acquire point cloud data of the machine body surface; The point cloud processing module is used to process point cloud data on the surface of the machine body. The point cloud reconstruction module is used to reconstruct the processed point cloud data to obtain a three-dimensional CAD model of the body surface. The target detection module is used to extract pit data from the 3D CAD model of the aircraft surface to obtain pit information; The pit identification module is used to determine the repairability of pits based on pit information, and to generate pit detection results on the machine surface and machine repair location information. The results output module is used to output and display the detection results of pits on the machine body surface and the machine body repair location information; The remote control module and the point cloud acquisition module exchange information via wireless communication. The point cloud processing module includes: a point cloud compensation unit, a noise processing unit, an adaptive repair unit, and a simplification processing unit; The point cloud compensation unit is used to compensate for atmospheric parameters in the point cloud data of the aircraft surface; The noise processing unit is used to filter, denoise, and smooth the point cloud data of the machine surface after compensation processing; The adaptive repair unit is used to perform B-spline interpolation adaptive repair on the noise-reduced point cloud data of the aircraft surface; The simplification processing unit is used to simplify the feature values ​​of the repaired body surface point cloud data, and the simplified body surface point cloud data is transmitted to the point cloud reconstruction module. The point cloud compensation unit specifically includes: an atmospheric temperature compensation module, an atmospheric pressure compensation module, an atmospheric humidity compensation module, and an airflow compensation module; Each compensation module sequentially performs temperature compensation, pressure compensation, humidity compensation, and airflow compensation on the point cloud data of the machine surface.

2. The body dent detection device based on point cloud data as described in claim 1, characterized in that, The point cloud acquisition module includes: a six-degree-of-freedom robotic arm, a monitoring console, a binocular camera, a decoder, and a point cloud acquisition card; The six-degree-of-freedom robotic arm includes: three translation servo motors and matching translation servo motor drivers, used to realize translation along the three coordinate axes OX, OY, and OZ; three rotation servo motors and matching rotation servo motor drivers, used to realize rotation along the three coordinate axes OX, OY, and OZ; the robotic arm is used to guide the monitoring station to move along the detection trajectory information. The servo motor driver is used to receive detection trajectory information of different aircraft models sent by the remote control module; A binocular camera, a decoder, and a point cloud acquisition card are installed on the monitoring console. The binocular camera performs six degrees of freedom transformations with the monitoring console to acquire point cloud data of the aircraft surface. The decoder is used to control the binocular camera to focus and obtain clear point cloud data of the aircraft surface. The point cloud acquisition card is used to store the point cloud data of the aircraft surface and to transmit the stored point cloud data of the aircraft surface to the point cloud processing module.

3. The body pit detection device based on point cloud data as described in claim 2, characterized in that, The point cloud acquisition module also includes: a brightness sensor, an optocoupler, a relay, and structured light; A brightness sensor is used to send the collected ambient brightness signal to a decoder, which generates a corresponding structured light control signal based on the ambient brightness signal. An optocoupler is used to receive the structured light control signal sent by the decoder and control a relay based on the structured light control signal. The relay is used to turn the structured light on or off and to adjust the brightness of the structured light. The structured light is used to illuminate the surface of the body facing the binocular camera.

4. The body dent detection device based on point cloud data as described in claim 1, characterized in that, The point cloud reconstruction module includes: a registration and stitching unit and a surface reconstruction unit; The registration and stitching unit performs registration and stitching processing on the simplified point cloud data of the body surface through a coordinate transformation matrix; The surface reconstruction unit uses the least squares method to reconstruct the surface of the aircraft surface point cloud data after registration and stitching, and obtains a three-dimensional CAD model of the aircraft surface.

5. The body dent detection device based on point cloud data as described in claim 1, characterized in that, The target detection module includes: a size unification unit, a pit detection unit, a pit data extraction unit, and a data output unit; The dimension unification unit is used to unify the dimensions of the 3D CAD model of the aircraft surface. The pit detection unit is used to identify and detect pits on the 3D CAD model of the body surface after the dimensions have been uniformly processed. The pit data extraction unit is used to extract data from each pit that has been identified and detected. The data extraction includes the extraction of the following pit information: the minimum width W of the pit, the depth Y at the width W, the ratio of the width W to the depth Y (i.e., the depth ratio), the structure name corresponding to the pit, and the part number corresponding to the pit. The data output unit is used to transmit the extracted pit information to the pit discrimination module.

6. The body dent detection device based on point cloud data as described in claim 1, characterized in that, The pit discrimination module includes: a pit classification unit, a pit determination unit, and a result loading unit; The pit classification unit is used to classify pits into two categories based on pit information: mandatory pits and pits to be determined. The pit determination unit is used to determine whether the pit to be determined is a pit that needs to be repaired; The result loading unit is used to summarize the information obtained synchronously with the point cloud data and correspond the information with the dents that must be repaired and the dents that need to be repaired; then it forms complete dent information by combining the location, section, part number and corresponding material of the dents on the machine body, and transmits the complete information to the result output module.

7. The body dent detection device based on point cloud data as described in claim 1, characterized in that, The result output module includes: an LED display screen, a power button, an alarm indicator light, and an alarm button; The LED display screen is used to display complete information about the dents; The power button is used to control the power supply to the LED display screen; The alarm indicator light is used to alert the user to dents that require repair or need to be repaired. The alarm button is used to turn off the alarm indicator light and sends the pressed status to the LED display screen for display.

8. A body dent detection system based on point cloud data, characterized in that, include: Based on the aircraft segment division location, each segment is equipped with a fuselage dent detection device based on point cloud data as described in any one of claims 1-7, in the following manner: A corresponding number of sliding rails are installed on the ground according to the number of sections; Each slide rail is equipped with a mounting frame. Two body dent detection devices based on point cloud data are installed in a set, with each set corresponding to a mounting frame. The mounting frame drives the body pit detection equipment based on point cloud data to move back and forth along the slide rail; One set of point cloud data-based fuselage dent detection equipment is used to detect the upper surface of aircraft sections, while the other set is used to detect the lower surface of aircraft sections.

9. A method for detecting body dents based on point cloud data, characterized in that, The specific steps are as follows: After the body pit detection device based on point cloud data as described in any one of claims 1-7 is powered on, it acquires point cloud data of the body surface through the point cloud acquisition module. The point cloud data of the machine surface acquired by the point cloud acquisition module is stored using a point cloud acquisition card; The atmospheric parameters of the body surface point cloud data stored in the storage unit are compensated by the point cloud compensation unit in the point cloud processing module. The noise processing unit in the point cloud processing module performs filtering, noise reduction, and smoothing processing on the body surface point cloud data after compensation by the point cloud compensation unit. The adaptive repair unit in the point cloud processing module performs B-spline interpolation adaptive repair on the noise-processed and smoothed point cloud data of the body surface by the noise processing unit. The feature value simplification process is performed on the point cloud data of the body surface repaired by the adaptive repair unit through the simplification processing unit in the point cloud processing module. The registration and stitching unit in the point cloud reconstruction module performs registration and stitching processing on the body surface point cloud data processed by the point cloud processing module. The surface reconstruction unit in the point cloud reconstruction module uses the least squares method to reconstruct the surface of the entire machine from the point cloud data of the registration and stitching unit. The size unification unit in the target detection module is used to unify the size of the 3D CAD model of the body after the point cloud reconstruction. The pit detection unit in the target detection module is used to identify and detect pits in the 3D CAD model after the dimensions are unified. The pit data extraction unit in the target detection module extracts the data of the identified pits; The pit data is transmitted to the pit discrimination module through the data output unit in the target detection module. The pit data obtained is classified by the pit classification unit in the pit discrimination module. The pit identification module uses a pit determination unit to determine the pits to be repaired and the location of the pits on the machine body that need repair, based on pit maintainability standards. The result loading unit in the dent identification module merges the location of the dents on the machine body to be repaired into complete information. The results obtained from the pit detection module are displayed through the result output module.

Images