Large-maneuvering high-overload wing deformation measurement method based on binocular vision

Abstract

The invention provides a large-maneuvering high-overload wing deformation measurement method based on binocular vision. The method comprises the following steps: setting wing surface feature identification points; calibrating a wing binocular camera and acquiring parameters of the wing binocular camera; acquiring a plurality of wing images in the large-maneuver high-overload flight process of the aircraft based on a wing binocular camera; and aiming at the wing image, calculating to obtain wing deformation data in the large-maneuver high-overload flight process of the aircraft. According to the method, the step change of the gray scale of the wing surface feature identification points between continuous frames is considered, a wing surface feature identification point identification model is constructed based on a frame difference method, and meanwhile, a wing surface feature identification point stereo matching model is constructed by virtue of binocular stereo vision; the method can overcome the defects in wing deformation measurement in a large-maneuverability and high-overload environment in the prior art, realizes high-precision, real-time and stable measurement of wing deformation, and has practical application value.

Description

Technical Field

[0001] The invention belongs to the technical field of aviation engineering measurement, and in particular to a method for measuring deformation of a large maneuverable high-overload wing based on binocular vision. Background Art

[0002] With the development of aviation technology, modern aircraft wings will be subjected to large aerodynamic loads and inertial forces during high-speed and large-maneuver flight, resulting in complex deformation of the wings. Accurately measuring wing deformation is extremely important for aircraft performance evaluation, structural optimization design, flight safety assurance, and aeroelasticity research.

[0003] Traditional wing deformation measurement methods, such as strain gauge measurement, can reflect the local strain of the wing to a certain extent, but they have the disadvantages of limited measurement points, difficulty in obtaining comprehensive wing deformation information, and certain intrusion into the wing structure. Optical measurement methods, such as laser interferometry, have high requirements for the measurement environment and poor stability in complex flight environments. As a non-contact optical measurement method, binocular vision measurement technology has the advantages of wide measurement range and the ability to obtain three-dimensional information. However, in high-maneuver and high-overload flight environments, it faces problems such as easy failure of camera calibration, reduced image matching accuracy, and difficulty in ensuring measurement accuracy. At present, there is no perfect solution that can effectively meet the actual engineering needs. Therefore, in order to overcome the shortcomings of the existing technology in the measurement of wing deformation in high-maneuver and high-overload environments, it is urgent and necessary to seek a high-maneuver and high-overload wing deformation measurement method based on binocular vision to achieve high-precision, real-time, and stable measurement of wing deformation in complex flight environments. Summary of the invention

[0004] In view of the defects in the above-mentioned prior art, the present invention proposes a method for measuring wing deformation under high maneuverability and high overload based on binocular vision. The method includes setting wing surface feature identification points; calibrating the wing binocular camera and obtaining the parameters of the wing binocular camera; obtaining a number of wing images during the high maneuverability and high overload flight of the aircraft based on the wing binocular camera; and calculating and obtaining the wing deformation data during the high maneuverability and high overload flight of the aircraft based on the wing images. The present invention takes into account the step change of the grayscale of the wing surface feature identification points between consecutive frames, constructs a wing surface feature identification point recognition model based on the frame difference method, and constructs a wing surface feature identification point stereo matching model with the help of binocular stereo vision. It can overcome the shortcomings of the prior art in the measurement of wing deformation under high maneuverability and high overload environment, and realize high-precision, real-time and stable measurement of wing deformation, which has practical application value.

[0005] The present invention provides a binocular vision-based large maneuver high-overload wing deformation measurement method, which comprises the following steps:

[0006] S1. Setting wing surface feature identification points: Setting a number of high reflectivity wing surface feature identification points in the wing to be tested area;

[0007] S2. Calibrate the wing binocular camera and obtain the parameters of the wing binocular camera: When the aircraft is in a stationary state, calibrate the wing binocular camera based on the Zhang Dingyou calibration method to obtain the parameters of the wing binocular camera;

[0008] S3. Acquiring several wing images during the high-maneuver and high-overload flight of the aircraft based on the wing binocular camera: During the flight of the aircraft, for the high-maneuver and high-overload flight phase of the aircraft, several wing images are synchronously captured and stored based on the wing binocular camera;

[0009] S31, installing the wing binocular camera on the aircraft body;

[0010] S32, determining the power supply source and data transmission of the wing binocular camera;

[0011] S33, considering the synchronous shooting requirements of the cameras, setting a wing master-slave camera: the wing master-slave camera includes a wing master camera and a wing slave camera, the wing master camera is provided with a synchronization signal generating module inside, and a synchronization pulse signal is generated by means of the synchronization signal generating module, and the synchronization pulse signal is sent to the wing slave camera via a connecting cable;

[0012] S34, the wing master and slave cameras synchronously shoot and acquire a number of wing images: during the flight of the aircraft, for the high-maneuver and high-overload flight stage of the aircraft, the wing slave camera starts to shoot a number of wing images synchronously with the wing master camera according to the received synchronization pulse signal;

[0013] S4. Calculate and obtain the wing deformation data of the aircraft during the high-maneuver and high-overload flight process for the wing image: with respect to the corresponding wing image obtained by the wing master and slave cameras, by means of wing image preprocessing and stereo matching of wing surface feature identification points, combined with the wing binocular camera parameters, set a deformation calculation module to calculate the three-dimensional coordinate changes of the wing surface feature identification points, and obtain the wing deformation data of the aircraft during the high-maneuver and high-overload flight process;

[0014] S41, by means of wing image preprocessing, respectively obtaining the wing foreground images of the wing master and slave cameras including the wing surface feature identification points: by means of the step change of the grayscale of the wing surface feature identification points between frames, constructing a wing surface feature identification point recognition model based on the frame difference method, and respectively performing preprocessing operations on the corresponding wing images obtained by the wing master and slave cameras through the wing surface feature identification point recognition model, and respectively obtaining the wing foreground images of the wing master and slave cameras including the wing surface feature identification points;

[0015] S42, using an edge operator to respectively obtain edge contours of wing surface feature identification points of the wing master and slave cameras: for the wing foreground images of the wing master and slave cameras, respectively, using an edge operator to extract edge contours of the wing surface feature identification points, and respectively obtain edge contours of the wing surface feature identification points of the wing master and slave cameras;

[0016] S43, using the least squares method, respectively obtaining the pixel coordinates of the center point of the wing surface feature identification point of the wing master and slave cameras: based on the pixel coordinates of all pixel points on the edge contour of the wing surface feature identification point of the wing master and slave cameras, respectively constructing the overdetermined equation group of the edge contour of the wing surface feature identification point of the wing master and slave cameras by using the least squares method, and combining with the singular value decomposition, respectively calculating and obtaining the pixel coordinates of the center point of the wing surface feature identification point of the wing master and slave cameras;

[0017] S44, obtaining the world coordinates of the center point of the wing surface feature identification point by means of stereo matching of the wing surface feature identification point: constructing a stereo matching model of the wing surface feature identification point based on the wing binocular camera parameters, and obtaining the world coordinates of the center point of the wing surface feature identification point based on the pixel coordinates of the center point of the wing surface feature identification point of the wing master and slave cameras through the stereo matching model of the wing surface feature identification point; wherein the stereo matching model of the wing surface feature identification point is:

[0018]

[0019]

[0020] Wherein, u1, v1 represent the pixel coordinates of the center point of the wing surface feature identification point on the horizontal axis and the vertical axis in the pixel coordinate system of the wing main camera respectively; u2, v2 represent the pixel coordinates of the center point of the wing surface feature identification point on the horizontal axis and the vertical axis in the pixel coordinate system of the wing slave camera respectively; Z C1 ,Z C2 They respectively represent the depth values ​​of the center points of the wing surface feature identification points in the camera coordinate systems of the wing main camera and the wing slave camera; dx1, dy1 respectively represent the physical size of each pixel in the intrinsic parameters of the wing main camera on the horizontal axis and the vertical axis; dx2, dy2 respectively represent the physical size of each pixel in the intrinsic parameters of the wing slave camera on the horizontal axis and the vertical axis; f x1 ,f y1 They represent the components of the focal length on the horizontal axis and the vertical axis in the intrinsic parameters of the wing main camera respectively; x2 ,f y2 They represent the components of the focal length of the wing on the horizontal and vertical axes in the intrinsic parameters of the camera respectively; W ,Y W ,Z WThey respectively represent the left-right, front-back, and vertical coordinates of the center point of the wing surface feature identification point in the world coordinate system; R and T respectively represent the rotation matrix and translation matrix of the wing binocular camera;

[0021] S45, setting a deformation calculation module to respectively calculate the three-dimensional coordinate changes of all wing surface feature identification points during the aircraft's high-maneuverability and high-overload flight, and obtaining the wing deformation data during the aircraft's high-maneuverability and high-overload flight.

[0022] Further, the construction of the wing surface feature identification point recognition model in step S41 specifically includes the following steps:

[0023] S411, considering that the background in two consecutive frames of the wing image is relatively still while the characteristic identification points on the wing surface change, the area where the characteristic identification points on the wing surface are located is displayed by using a two-frame difference method to obtain a wing difference image D t (x,y):

[0024] D t (x,y)=I t (x,y)-I t-1 (x,y) (1)

[0025] Among them, I t (x,y),I t-1 (x, y) represent the wing image frames corresponding to time t and time t-1 respectively; x, y represent the pixel coordinates of the horizontal axis and the vertical axis in the wing image respectively;

[0026] S412, setting a wing surface feature threshold T, for the wing differential image D t (x,y) is binarized to obtain the wing foreground image f t (x,y):

[0027]

[0028] S413, taking the wing foreground image f t (x, y), the wing foreground image f is displayed by morphological filtering t The wing surface feature points in (x, y) are identified and the clutter is removed.

[0029] Preferably, the step S1 specifically includes the following steps:

[0030] S11. Determine the wing test area: Consider the structural characteristics and key deformation parts of the wing, and determine the wing test area where the wing surface feature identification points need to be set;

[0031] S12, making characteristic marking points on the wing surface: considering the light environment during the flight of the aircraft, selecting materials with high reflectivity to make characteristic marking points on the wing surface;

[0032] S13, pasting wing surface feature identification points: considering the material of the wing surface feature identification points and the characteristics of the wing surface, pasting the wing surface feature identification points on the wing to be tested area;

[0033] The step S2 specifically includes the following steps:

[0034] S21, determining a calibration plate for calibrating the wing binocular camera: considering the specific structural characteristics and shape of the wing, a chessboard of corresponding size is selected as a calibration plate for calibrating the wing binocular camera;

[0035] S22, obtaining several images of the calibration plate for calibrating the wing binocular camera: considering different angles and distances, according to the specific structure and shape of the wing, moving the calibration plate for calibrating the wing binocular camera, and photographing and obtaining several images of the calibration plate for calibrating the wing binocular camera;

[0036] S23, solving the initial values ​​of the intrinsic parameters of the wing binocular camera: based on the images of the calibration plates used for calibration of the wing binocular cameras, constructing the corresponding homography matrices of the wing respectively, and solving the initial values ​​of the intrinsic parameters of the wing binocular camera according to the Zhang Zhengyou calibration method;

[0037] S24, optimizing the intrinsic parameters of the wing binocular camera: optimizing and adjusting the initial values ​​of the intrinsic parameters of the wing binocular camera by means of a nonlinear least squares method to obtain optimized intrinsic parameters of the wing binocular camera;

[0038] S25. Solving the extrinsic parameters of the wing binocular camera: Solving the extrinsic parameters of the wing binocular camera based on the optimized intrinsic parameters of the wing binocular camera and the constructed plurality of wing homography matrices, combined with the geometric relationship of the wing binocular camera; the extrinsic parameters of the wing binocular camera include the rotation matrix R and the translation matrix T of the wing binocular camera.

[0039] Preferably, the step S31 specifically includes the following steps:

[0040] S311. Considering the impact and vibration caused by high overload of large maneuvers of the aircraft, prepare a mounting bracket with high strength and good shock absorption performance;

[0041] S312, installing the wing binocular camera at a specific position of the aircraft body by means of the mounting bracket, wherein the specific position of the aircraft body takes into account that the field of view of the wing binocular camera can completely cover the wing area to be measured;

[0042] The step S32 specifically includes the following steps:

[0043] S321, considering the power consumption requirement of the wing binocular camera during the entire flight of the aircraft, connecting the power supply port of the wing binocular camera to the power supply system of the aircraft;

[0044] S322. Considering the data transmission and storage requirements of the wing binocular camera during the entire flight of the aircraft, the data output port of the wing binocular camera is connected to the aircraft's onboard data storage device via a data transmission line.

[0045] Preferably, the synchronization signal generating module in the step S33 generates a synchronization pulse signal based on its own high-precision clock signal; in the step S34, during the shooting process, the wing master-slave cameras continue to maintain a synchronization state, and each time a synchronization pulse signal is received, the wing master-slave cameras simultaneously perform a wing image acquisition.

[0046] Preferably, the wing binocular camera parameters in step S2 include wing binocular camera internal parameters and wing binocular camera external parameters.

[0047] Compared with the prior art, the technical effects of the present invention are:

[0048] 1. The binocular vision-based large maneuver and high-overload wing deformation measurement method designed by the present invention takes into account the step change of the grayscale of the wing surface feature identification points between consecutive frames, and constructs a wing surface feature identification point recognition model based on the frame difference method, thereby realizing efficient and high-precision extraction of the wing surface feature identification points; with the help of binocular stereo vision, a stereo matching model of the wing surface feature identification points is constructed, which effectively guarantees the effective measurement of the wing deformation under large maneuvers and high overloads.

[0049] 2. The binocular vision-based large-maneuver and high-overload wing deformation measurement method designed by the present invention has good adaptability, can overcome the shortcomings of the prior art in the measurement of wing deformation under large-maneuver and high-overload environments, and achieve high-precision, real-time and stable measurement of wing deformation, so as to effectively meet actual engineering needs, provide reliable technical support for the wing deformation measurement of aircraft, and play an important role in improving aircraft performance and ensuring flight safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Other features, objects and advantages of the present application will become more apparent from the detailed description of non-limiting embodiments made with reference to the following drawings.

[0051] Figure 1 It is a flow chart of the method for measuring deformation of a large maneuvering high-overload wing based on binocular vision of the present invention;

[0052] Figure 2 is a schematic diagram of a coordinate system constructed by the wing binocular camera of the present invention;

[0053] Figure 3 It is a schematic diagram of a flow chart of acquiring wing deformation data during high-maneuver and high-overload flight of an aircraft according to the present invention. DETAILED DESCRIPTION

[0054] The present application will be further described in detail below in conjunction with the accompanying drawings and embodiments. It will be appreciated that the specific embodiments described herein are only used to explain the relevant inventions, rather than to limit the inventions. It should also be noted that, for ease of description, only portions related to the relevant inventions are shown in the accompanying drawings. It should be noted that, in the absence of conflict, the embodiments in the present application and the features in the embodiments may be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in conjunction with the embodiments.

[0055] Figure 1 The present invention shows a binocular vision-based large maneuver high overload wing deformation measurement method. The term large maneuver in the present invention refers to the measurement speed range: M = 0 ~ 1.3M; high overload refers to the overload range: <6g. The method of the present invention includes the following steps:

[0056] S1. Setting wing surface feature identification points: Setting a number of high reflectivity wing surface feature identification points in the wing to be tested area.

[0057] S11. Determine the wing test area: Consider the structural characteristics of the wing and the key deformation parts, and determine the wing test area where the wing surface feature identification points need to be set.

[0058] S12. Making wing surface feature marking points: Considering the lighting environment during the flight of the aircraft, select materials with high reflectivity to make the wing surface feature marking points to ensure that under different lighting conditions, especially the lighting environment during the flight of the aircraft, the camera can clearly capture the wing surface feature marking points. The high reflectivity material refers to: mainly flexible reflective material, the reflective film adopts an anti-reflection film layer design, and the reflectivity is ≥85%.

[0059] S13. Paste the wing surface feature marking points: Consider the material of the wing surface feature marking points and the characteristics of the wing surface, and paste the wing surface feature marking points on the wing to be tested area to avoid bubbles or wrinkles and ensure that the sticker fits tightly to the surface.

[0060] S2. Calibrate the wing binocular camera and obtain the wing binocular camera parameters: When the aircraft is stationary, calibrate the wing binocular camera based on the Zhang Dingyou calibration method to obtain the wing binocular camera parameters. The wing binocular camera parameters include the wing binocular camera internal parameters and the wing binocular camera external parameters.

[0061] S21. Determine the calibration plate for calibrating the wing binocular camera: Considering the specific structural characteristics and shape of the wing, select a chessboard of corresponding size as the calibration plate for calibrating the wing binocular camera.

[0062] S22. Acquire several images of the calibration plate used for calibrating the wing binocular camera: Consider different angles and distances, and according to the specific structure and shape of the wing, move the calibration plate used for calibrating the wing binocular camera, and take pictures to acquire several images of the calibration plate used for calibrating the wing binocular camera.

[0063] S23. Solving the initial values ​​of the intrinsic parameters of the wing binocular camera: Based on several calibration plate images used for calibrating the wing binocular camera, construct several corresponding wing homography matrices respectively, and solve the initial values ​​of the intrinsic parameters of the wing binocular camera according to the Zhang Zhengyou calibration method.

[0064] S24, optimizing the intrinsic parameters of the wing binocular camera: optimizing and adjusting the initial values ​​of the intrinsic parameters of the wing binocular camera by means of a nonlinear least squares method to obtain optimized intrinsic parameters of the wing binocular camera.

[0065] S25. Solving the extrinsic parameters of the wing binocular camera: Based on the optimized intrinsic parameters of the wing binocular camera and several constructed wing homography matrices, combined with the geometric relationship of the wing binocular camera, the extrinsic parameters of the wing binocular camera are solved; the extrinsic parameters of the wing binocular camera include the rotation matrix R and translation matrix T of the wing binocular camera.

[0066] S3. Acquire several wing images during the large-maneuver and high-overload flight of the aircraft based on the wing binocular camera: During the flight of the aircraft, for the large-maneuver and high-overload flight phase of the aircraft, several wing images are synchronously captured and stored based on the wing binocular camera.

[0067] S31. Set a wing binocular camera on the aircraft body.

[0068] S311. Considering the impact and vibration caused by large maneuvers and high overload of the aircraft, prepare a mounting bracket with high strength and good shock absorption performance.

[0069] S312. Install the wing binocular camera at a specific position of the aircraft body with the aid of a mounting bracket. The specific position of the aircraft body takes into account that the field of view of the wing binocular camera can completely cover the wing area to be measured.

[0070] S32. Determine the power supply source and data transmission of the wing binocular camera.

[0071] S321. Considering the power consumption requirement of the wing binocular camera during the entire flight of the aircraft, the power supply port of the wing binocular camera is connected to the power supply system of the aircraft.

[0072] S322. Considering the data transmission and storage requirements of the wing binocular camera during the entire flight of the aircraft, the data output port of the wing binocular camera is connected to the aircraft's onboard data storage device via a data transmission line.

[0073] S33. Considering the synchronous shooting requirements of the camera, a wing master-slave camera is set: the wing master-slave camera includes a wing master camera and a wing slave camera. A synchronization signal generation module is provided inside the wing master camera. The synchronization signal generation module generates a synchronization pulse signal, and the synchronization pulse signal is sent to the wing slave camera via a connecting cable. The synchronization signal generation module generates a synchronization pulse signal based on its own high-precision clock signal.

[0074] In a specific embodiment, the coordinate system constructed by the wing binocular camera is as follows: Figure 2 As shown, C1 and C2 are the wing main camera and the wing slave camera, and the convergence point of their optical axes intersects with the Y axis of the space coordinate system, the intersection point is O, L is the baseline length, and H is the distance between the polar line and the center of the wing to be measured area.

[0075] S34. The wing master and slave cameras synchronously shoot and acquire several wing images: During the flight of the aircraft, for the high-maneuver and high-overload flight stage of the aircraft, the wing slave camera starts to shoot several wing images synchronously with the wing master camera based on the received synchronization pulse signal. During the shooting process, the wing master and slave cameras continue to maintain a synchronized state. Each time a synchronization pulse signal is received, the wing master and slave cameras simultaneously collect a wing image, thereby ensuring that each set of wing images obtained is taken at the same time, and can accurately reflect the actual deformation of the wing at each moment during the high-maneuver and high-overload flight process.

[0076] S4. Calculate and obtain the wing deformation data of the aircraft during high maneuverability and high overload flight for the wing image: For the corresponding wing images obtained by the wing master and slave cameras, with the help of wing image preprocessing and wing surface feature identification point stereo matching, combined with the wing binocular camera parameters, set the deformation calculation module to calculate the three-dimensional coordinate changes of the wing surface feature identification points, and obtain the wing deformation data of the aircraft during high maneuverability and high overload flight, such as Figure 3 shown.

[0077] S41. By means of wing image preprocessing, the wing foreground images of the wing master and slave cameras including the wing surface feature identification points are respectively obtained: by means of the step change of the grayscale of the wing surface feature identification points between frames, a wing surface feature identification point recognition model is constructed based on the frame difference method, and the corresponding wing images obtained by the wing master and slave cameras are respectively preprocessed by the wing surface feature identification point recognition model to obtain the wing foreground images of the wing master and slave cameras including the wing surface feature identification points. In a specific embodiment, the use scenario is an airborne test environment, and strong light interference is serious. The energy of the fill light after being reflected by the wing surface feature identification points is submerged in the sunlight, and it is extremely difficult to identify the wing surface feature identification points only by light intensity. The frame difference method uses the image dislocation subtraction processing technology, and by means of the step change of the target grayscale, the wing surface feature identification points and the background strong light are distinguished, which can effectively improve the detection contrast.

[0078] The construction of the wing surface feature identification point recognition model specifically includes the following steps:

[0079] S411, considering that the background in two consecutive frames of wing images is relatively static while the characteristic identification points on the wing surface change, the area where the characteristic identification points on the wing surface are located is displayed by using the two-frame difference method to obtain the wing difference image D t (x,y):

[0080] D t (x,y)=I t (x,y)-I t-1 (x,y) (1)

[0081] Among them, I t (x,y),I t-1 (x, y) represents the wing image frames corresponding to time t and time t-1 respectively; x, y represent the pixel coordinates of the horizontal axis and vertical axis in the wing image respectively.

[0082] S412, setting the wing surface feature threshold T, for the wing difference image D t (x, y) is binarized and D is judged t The size relationship between (x, y) and T is used to separate the foreground and background, and obtain the wing foreground image f t (x,y):

[0083]

[0084] S413, for taking the wing foreground image f t (x,y), using morphological filtering to display the wing foreground image f t Identify the wing surface feature points in (x,y) and remove the noise.

[0085] S42, using edge operators to respectively obtain edge contours of wing surface feature identification points of the wing master and slave cameras: for the wing foreground images of the wing master and slave cameras, using edge operators, edge contours of wing surface feature identification points are extracted to respectively obtain edge contours of wing surface feature identification points of the wing master and slave cameras. In a specific embodiment, the edge operator uses methods such as a canny operator and a Laplace operator.

[0086] S43. With the help of least squares method, the pixel coordinates of the center point of the wing surface feature identification point of the wing master and slave cameras are respectively obtained: based on the pixel coordinates of all pixel points on the edge contour of the wing surface feature identification points of the wing master and slave cameras, the overdetermined equation group of the edge contour of the wing surface feature identification points of the wing master and slave cameras is respectively constructed with the help of least squares method, and combined with singular value decomposition, the pixel coordinates of the center point of the wing surface feature identification point of the wing master and slave cameras are respectively calculated.

[0087] S44, obtaining the world coordinates of the center point of the wing surface feature identification point by means of stereo matching of the wing surface feature identification point: constructing a stereo matching model of the wing surface feature identification point based on the wing binocular camera parameters, and obtaining the world coordinates of the center point of the wing surface feature identification point based on the pixel coordinates of the center point of the wing surface feature identification point of the wing master and slave cameras through the wing surface feature identification point stereo matching model; the wing surface feature identification point stereo matching model is:

[0088]

[0089]

[0090] Among them, u1, v1 represent the pixel coordinates of the center point of the wing surface feature identification point on the horizontal axis and the vertical axis in the pixel coordinate system of the wing main camera respectively; u2, v2 represent the pixel coordinates of the center point of the wing surface feature identification point on the horizontal axis and the vertical axis in the pixel coordinate system of the wing slave camera respectively; Z C1 ,Z C2 They represent the depth values ​​of the center points of the wing surface feature identification points in the camera coordinate systems of the wing main camera and the wing slave camera respectively; dx1 and dy1 represent the physical size of each pixel in the intrinsic parameters of the wing main camera on the horizontal axis and the vertical axis respectively; dx2 and dy2 represent the physical size of each pixel in the intrinsic parameters of the wing slave camera on the horizontal axis and the vertical axis respectively; f x1 ,f y1 They represent the components of the focal length on the horizontal and vertical axes in the intrinsic parameters of the wing main camera respectively; f x2 ,f y2 They represent the components of the focal length of the wing from the intrinsic parameters of the camera on the horizontal and vertical axes respectively; W ,Y W ,Z WThey respectively represent the left-right, front-back and vertical coordinate values ​​of the center point of the wing surface feature identification point in the world coordinate system.

[0091] S45, setting a deformation calculation module to respectively calculate the three-dimensional coordinate changes of all wing surface feature identification points during the aircraft's high-maneuverability and high-overload flight, and obtaining the wing deformation data during the aircraft's high-maneuverability and high-overload flight.

[0092] The present invention designs a large maneuver and high overload wing deformation measurement method based on binocular vision. Taking into account the step change of grayscale of wing surface feature identification points between consecutive frames, a wing surface feature identification point recognition model is constructed based on the frame difference method, thereby realizing efficient and high-precision extraction of wing surface feature identification points; with the help of binocular stereo vision, a stereo matching model of wing surface feature identification points is constructed, which effectively guarantees the effective measurement of wing deformation under large maneuver and high overload; the proposed method has good adaptability, can overcome the shortcomings of the prior art in wing deformation measurement under large maneuver and high overload environment, and realize high-precision, real-time and stable measurement of wing deformation, so as to effectively meet actual engineering needs, provide reliable technical support for wing deformation measurement of aircraft, and play an important role in improving aircraft performance and ensuring flight safety.

[0093] Finally, it should be noted that the above embodiments are only intended to illustrate rather than limit the technical solutions of the present invention. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that the present invention can still be modified or replaced by equivalents. Any modification or partial replacement that does not depart from the spirit and scope of the present invention should be included in the scope of the claims of the present invention.

Claims

1. A method for measuring deformation of a high-maneuver high-overload wing based on binocular vision, characterized in that: It includes the following steps: S1. Setting wing surface feature identification points: Setting a number of high reflectivity wing surface feature identification points in the wing to be tested area; S2. Calibrate the wing binocular camera and obtain the parameters of the wing binocular camera: When the aircraft is in a stationary state, calibrate the wing binocular camera based on the Zhang Dingyou calibration method to obtain the parameters of the wing binocular camera; S3. Acquiring several wing images during the high-maneuver and high-overload flight of the aircraft based on the wing binocular camera: During the flight of the aircraft, for the high-maneuver and high-overload flight phase of the aircraft, several wing images are synchronously captured and stored based on the wing binocular camera; S31, installing the wing binocular camera on the aircraft body; S32, determining the power supply source and data transmission of the wing binocular camera; S33, considering the synchronous shooting requirements of the cameras, setting a wing master-slave camera: the wing master-slave camera includes a wing master camera and a wing slave camera, the wing master camera is provided with a synchronization signal generating module inside, and a synchronization pulse signal is generated by means of the synchronization signal generating module, and the synchronization pulse signal is sent to the wing slave camera via a connecting cable; S34, the wing master and slave cameras synchronously shoot and acquire a number of wing images: during the flight of the aircraft, for the high-maneuver and high-overload flight stage of the aircraft, the wing slave camera starts to shoot a number of wing images synchronously with the wing master camera according to the received synchronization pulse signal; S4. Calculate and obtain the wing deformation data of the aircraft during the high-maneuver and high-overload flight process for the wing image: with respect to the corresponding wing image obtained by the wing master and slave cameras, by means of wing image preprocessing and stereo matching of wing surface feature identification points, combined with the wing binocular camera parameters, set a deformation calculation module to calculate the three-dimensional coordinate changes of the wing surface feature identification points, and obtain the wing deformation data of the aircraft during the high-maneuver and high-overload flight process; S41, by means of wing image preprocessing, respectively obtaining the wing foreground images of the wing master and slave cameras including the wing surface feature identification points: by means of the step change of the grayscale of the wing surface feature identification points between frames, constructing a wing surface feature identification point recognition model based on the frame difference method, and respectively performing preprocessing operations on the corresponding wing images obtained by the wing master and slave cameras through the wing surface feature identification point recognition model, and respectively obtaining the wing foreground images of the wing master and slave cameras including the wing surface feature identification points; S42, using an edge operator to respectively obtain edge contours of wing surface feature identification points of the wing master and slave cameras: for the wing foreground images of the wing master and slave cameras, respectively, using an edge operator to extract edge contours of the wing surface feature identification points, and respectively obtain edge contours of the wing surface feature identification points of the wing master and slave cameras; S43, using the least squares method, respectively obtaining the pixel coordinates of the center point of the wing surface feature identification point of the wing master and slave cameras: based on the pixel coordinates of all pixel points on the edge contour of the wing surface feature identification point of the wing master and slave cameras, respectively constructing the overdetermined equation group of the edge contour of the wing surface feature identification point of the wing master and slave cameras by using the least squares method, and combining with the singular value decomposition, respectively calculating and obtaining the pixel coordinates of the center point of the wing surface feature identification point of the wing master and slave cameras; S44, obtaining the world coordinates of the center point of the wing surface feature identification point by means of stereo matching of the wing surface feature identification point: constructing a stereo matching model of the wing surface feature identification point based on the wing binocular camera parameters, and obtaining the world coordinates of the center point of the wing surface feature identification point based on the pixel coordinates of the center point of the wing surface feature identification point of the wing master and slave cameras through the stereo matching model of the wing surface feature identification point; wherein the stereo matching model of the wing surface feature identification point is: Wherein, u1, v1 represent the pixel coordinates of the center point of the wing surface feature identification point on the horizontal axis and the vertical axis in the pixel coordinate system of the wing main camera respectively; u2, v2 represent the pixel coordinates of the center point of the wing surface feature identification point on the horizontal axis and the vertical axis in the pixel coordinate system of the wing slave camera respectively; Z C1 ,Z C2 They respectively represent the depth values ​​of the center points of the wing surface feature identification points in the camera coordinate systems of the wing main camera and the wing slave camera; dx1, dy1 respectively represent the physical size of each pixel in the intrinsic parameters of the wing main camera on the horizontal axis and the vertical axis; dx2, dy2 respectively represent the physical size of each pixel in the intrinsic parameters of the wing slave camera on the horizontal axis and the vertical axis; f x1 ,f y1 They represent the components of the focal length on the horizontal axis and the vertical axis in the intrinsic parameters of the wing main camera respectively; x2 ,f y2 They represent the components of the focal length of the wing on the horizontal and vertical axes in the intrinsic parameters of the camera respectively; W ,Y W ,Z W They respectively represent the left-right, front-back, and vertical coordinates of the center point of the wing surface feature identification point in the world coordinate system; R and T respectively represent the rotation matrix and translation matrix of the wing binocular camera; S45, setting a deformation calculation module to respectively calculate the three-dimensional coordinate changes of all wing surface feature identification points during the aircraft's high-maneuverability and high-overload flight, and obtain the wing deformation data during the aircraft's high-maneuverability and high-overload flight.

2. The binocular vision-based large maneuver high-overload wing deformation measurement method according to claim 1 is characterized in that: The construction of the wing surface feature identification point recognition model in step S41 specifically includes the following steps: S411, considering that the background in two consecutive frames of wing images is relatively still while the characteristic identification points on the wing surface change, the area where the characteristic identification points on the wing surface are located is displayed by using a two-frame difference method to obtain a wing difference image D t (x,y): D t (x,y)=I t (x,y)-I t-1 (x,y) (1) Among them, I t (x,y),I t-1 (x, y) represent the wing image frames corresponding to time t and time t-1 respectively; x, y represent the pixel coordinates of the horizontal axis and the vertical axis in the wing image respectively; S412, setting a wing surface feature threshold T, for the wing differential image D t (x,y) is binarized to obtain the wing foreground image f t (x,y): S413, taking the wing foreground image f t (x, y), the wing foreground image f is displayed by morphological filtering t The wing surface feature points in (x, y) are identified and the clutter is removed.

3. The method for measuring wing deformation under high maneuverability and high overload based on binocular vision according to claim 1 is characterized in that: The step S1 specifically includes the following steps: S11. Determine the wing test area: Consider the structural characteristics and key deformation parts of the wing, and determine the wing test area where the wing surface feature identification points need to be set; S12, making characteristic marking points on the wing surface: considering the light environment during the flight of the aircraft, selecting materials with high reflectivity to make characteristic marking points on the wing surface; S13, pasting wing surface feature identification points: considering the material of the wing surface feature identification points and the characteristics of the wing surface, pasting the wing surface feature identification points on the wing to be tested area; The step S2 specifically includes the following steps: S21, determining a calibration plate for calibrating the wing binocular camera: considering the specific structural characteristics and shape of the wing, a chessboard of corresponding size is selected as a calibration plate for calibrating the wing binocular camera; S22, obtaining several images of the calibration plate for calibrating the wing binocular camera: considering different angles and distances, according to the specific structure and shape of the wing, moving the calibration plate for calibrating the wing binocular camera, and photographing and obtaining several images of the calibration plate for calibrating the wing binocular camera; S23, solving the initial values ​​of the intrinsic parameters of the wing binocular camera: based on the images of the calibration plates used for calibration of the wing binocular cameras, constructing the corresponding homography matrices of the wing respectively, and solving the initial values ​​of the intrinsic parameters of the wing binocular camera according to the Zhang Zhengyou calibration method; S24, optimizing the intrinsic parameters of the wing binocular camera: optimizing and adjusting the initial values ​​of the intrinsic parameters of the wing binocular camera by means of a nonlinear least squares method to obtain optimized intrinsic parameters of the wing binocular camera; S25. Solving the extrinsic parameters of the wing binocular camera: Solving the extrinsic parameters of the wing binocular camera based on the optimized intrinsic parameters of the wing binocular camera and the constructed plurality of wing homography matrices, combined with the geometric relationship of the wing binocular camera; the extrinsic parameters of the wing binocular camera include the rotation matrix R and the translation matrix T of the wing binocular camera.

4. The method for measuring wing deformation under high maneuverability and high overload based on binocular vision according to claim 1 is characterized in that: The step S31 specifically includes the following steps: S311. Considering the impact and vibration caused by high overload caused by large maneuvers of the aircraft, prepare the mounting bracket; S312, installing the wing binocular camera at a specific position of the aircraft body by means of the mounting bracket, wherein the specific position of the aircraft body takes into account that the field of view of the wing binocular camera can completely cover the wing area to be measured; The step S32 specifically includes the following steps: S321, considering the power consumption requirement of the wing binocular camera during the entire flight of the aircraft, connecting the power supply port of the wing binocular camera to the power supply system of the aircraft; S322. Considering the data transmission and storage requirements of the wing binocular camera during the entire flight of the aircraft, the data output port of the wing binocular camera is connected to the aircraft's onboard data storage device via a data transmission line.

5. The method for measuring wing deformation under high maneuverability and high overload based on binocular vision according to claim 1 is characterized in that: In step S33, the synchronization signal generating module generates a synchronization pulse signal based on its own clock signal; During the shooting process in step S34, the wing master-slave cameras continue to maintain a synchronized state, and each time a synchronization pulse signal is received, the wing master-slave cameras simultaneously perform a wing image acquisition.

6. The binocular vision-based large maneuver high overload wing deformation measurement method according to claim 1 is characterized in that: The wing binocular camera parameters in step S2 include wing binocular camera internal parameters and wing binocular camera external parameters.

Images