Plane conversion method, device and system taking crack plane as reference coordinate system

Abstract

The invention discloses a plane conversion method taking a crack plane as a reference coordinate system, which comprises the following steps of: determining a crack space coordinate system through a camera attitude when a first crack image is shot, and respectively representing projection point space coordinates corresponding to four vertexes on a camera frame through a space relative relationship between the crack plane and a camera plane; establishing a camera coordinate system, and calculating the spatial relative relationship; substituting the space relative relation into a crack space coordinate system to obtain projection point space coordinates corresponding to the four vertexes; establishing a camera coordinate system for each crack image shot subsequently, calculating the absolute scale of each crack image, and correcting the crack image based on the absolute scale; and splicing the plurality of corrected crack images through feature matching, and converting the crack images into a crack space coordinate system through a relative relation between the spliced crack images. On the basis, plane conversion between the camera coordinate system and the crack plane coordinate system is more efficiently and accurately realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of image processing and computer vision, and particularly relates to a plane conversion method, device and system taking a crack plane as a reference coordinate system. BACKGROUND

[0002] Traditional manual detection methods have strong subjectivity, high work intensity, low efficiency and safety hazards, and cannot meet the needs of engineering practice. Although the crack detection methods based on threshold, edge, region, matching and other digital image processing technologies improve objectivity, the process is complex and easily affected by the environment. Existing crack detection technologies, such as binocular vision solutions, have good scene adaptability, but have high hardware costs and complex calculations. Monocular vision technology has an advantage in hardware cost compared to binocular vision solutions, but needs to solve the problem of how to accurately measure the physical parameters of the crack. With the development of deep learning technology, the research in the field of crack detection gradually changes from traditional image processing technology to deep learning-based methods. Deep learning algorithms can automatically extract crack features to realize crack recognition and extraction, but the image processing process is complex and easily affected by the environment.

[0003] In existing crack detection technologies, the conversion between the camera coordinate system and the crack plane coordinate system has the problems of high complexity and insufficient accuracy. Traditional methods rely on a single perspective or simple projection, making it difficult to achieve efficient image stitching and accurate three-dimensional coordinate calculation. In addition, the existing technology is not perfect in modeling the geometric relationship of the crack plane, resulting in large reconstruction result errors. SUMMARY

[0004] To overcome the problems of high complexity and poor accuracy in the conversion between the camera coordinate system and the crack plane coordinate system in the existing technology, the present application provides a plane conversion method, device and system taking a crack plane as a reference coordinate system, which establishes a space coordinate system based on the crack plane and combines infrared projection and geometric parameter calculation to make the plane conversion between the camera coordinate system and the crack plane coordinate system more efficient and accurate.

[0005] According to an aspect of the present disclosure, a plane conversion method based on a crack plane coordinate system is provided, including: determining a crack space coordinate system through a camera pose when a first crack image is captured, and representing the space coordinates of each point in the crack space coordinate system through a space relative relationship between the crack plane and the camera plane; establishing a camera coordinate system with a vertex on the camera frame as the origin, and calculating the space relative relationship between the crack plane and the camera plane in the camera coordinate system; substituting the space relative relationship between the crack plane and the camera plane into the crack space coordinate system to obtain the space coordinates of each point in the crack space coordinate system; establishing a camera coordinate system for each subsequent crack image, calculating the absolute scale of each crack image, and performing correction processing on the crack image based on the absolute scale; and splicing multiple corrected crack images through feature matching in the crack plane coordinate system, and converting to the crack space coordinate system through the relative relationship between the spliced crack images.

[0006] Further, the relative position between the camera and the crack plane when the first crack image is captured is determined as the y-axis; wherein the projection point of the camera on the crack plane when the first crack image is captured is the origin; a parallel line of the horizontal frame of the camera frame passing through the origin on the crack plane is the x-axis; and a vertical line of the x-axis passing through the origin on the crack plane is the z-axis, to establish the crack space coordinate system.

[0007] Further, representing the space coordinates of each point in the crack space coordinate system through the space relative relationship between the crack plane and the camera plane includes: representing the space coordinates of each point on the camera plane through the space relative relationship between the crack plane and the camera plane; calculating a plane normal vector of the camera plane, and representing the straight line equation of the projection line of each point on the camera plane relative to the crack plane in the crack space coordinate system based on the plane normal vector and the space coordinates of each point on the camera plane; and representing the space coordinates of the projection point of each point on the camera plane on the crack plane based on the straight line equation.

[0008] Further, a camera coordinate system is established with a vertex in the camera frame as the origin, a parallel line of the horizontal frame of the camera frame passing through the origin on the camera plane as the x-axis, a parallel line of the vertical frame of the camera frame passing through the origin on the camera plane as the z-axis, and a line perpendicular to the camera plane and passing through the origin as the y-axis.

[0009] Further, the spatial relative relationship between the camera plane and the crack plane is calculated in the camera coordinate system, including: calculating a deflection angle of the camera plane on the x-axis relative to the crack plane based on the camera frame side length and the four distances measured by the distance meter; calculating a deflection angle of the camera plane on the z-axis relative to the crack plane based on the camera frame side length and the four distances measured by the distance meter; and the spatial relative relationship includes the deflection angles of the camera plane on the x-axis and the z-axis relative to the crack plane.

[0010] Further, the absolute scale of each crack image is calculated, including: calculating the four side lengths and four corners of the projection plane of the camera frame on the crack plane based on the camera frame side length and the four distances measured by the distance meter; and obtaining the absolute scale of the crack image based on the four side lengths and four corners of the projection plane.

[0011] Further, the crack image is corrected based on the absolute scale, including: obtaining the undistorted coordinates of each pixel point in the crack image; obtaining the pixel value at the undistorted coordinates from the crack image before correction, and assigning the pixel value to the pixel point position in the crack image after correction through linear interpolation; establishing a proportional relationship between the image pixel and the actual size based on the absolute scale; and converting the coordinates of each pixel point in the crack image into actual physical coordinates based on the proportional relationship.

[0012] According to an aspect of the present application, a plane conversion device with a crack plane as a reference coordinate system is provided, including: a monocular crack recognition instrument for detecting cracks and, when a crack is detected, capturing a crack image and measuring distance information between the crack plane; and a terminal for obtaining the crack image and the distance information from the monocular crack recognition instrument, and executing the plane conversion method with the crack plane as the reference coordinate system.

[0013] According to an aspect of the present application, a plane conversion system with a crack plane as a reference coordinate system is provided, comprising: a crack space coordinate system establishing module, configured to determine a crack space coordinate system through a camera pose when a first crack image is captured, and express space coordinates of each point in the crack space coordinate system through a space relative relationship between a crack plane and a camera plane; a camera coordinate system establishing module, configured to establish a camera coordinate system with one vertex on a camera frame as an origin, and calculate the space relative relationship between the crack plane and the camera plane in the camera coordinate system; a plane conversion module, configured to substitute the space relative relationship between the crack plane and the camera plane into the crack space coordinate system to obtain the space coordinates of each point in the crack space coordinate system; a crack image correction module, configured to establish a camera coordinate system for each subsequent crack image, calculate an absolute scale of each crack image, and correct the crack image based on the absolute scale; and a crack image stitching module, configured to stitch multiple corrected crack images through feature matching in a crack plane coordinate system, and convert to a crack space coordinate system through a relative relationship between the stitched crack images.

[0014] According to an aspect of the present application, a non-transitory computer readable storage medium is provided, which stores computer instructions, and the computer instructions cause the computer to perform the plane conversion method with a crack plane as a reference coordinate system.

[0015] The above technical solution determines a crack space coordinate system through a camera pose when a first crack image is captured, expresses space coordinates of four projection points corresponding to four vertices on a camera frame in the crack space coordinate system through a space relative relationship between a crack plane and a camera plane, establishes a camera coordinate system with one vertex on the camera frame as an origin, calculates the space relative relationship between the crack plane and the camera plane in the camera coordinate system, substitutes the space relative relationship between the crack plane and the camera plane into the crack space coordinate system to obtain the space coordinates of the four projection points corresponding to the four vertices on the camera frame in the crack space coordinate system, establishes a camera coordinate system for each subsequent crack image, calculates an absolute scale of each crack image, and corrects the crack image based on the absolute scale, and stitches multiple corrected crack images through feature matching in a crack plane coordinate system, and converts to a crack space coordinate system through a relative relationship between the stitched crack images.

[0016] Compared with the prior art, the present application has the following beneficial effects:

[0017] (1) The present application more efficiently and accurately realizes plane conversion between a camera coordinate system and a crack plane coordinate system by establishing a crack space coordinate system and calculating infrared projection point coordinates and normal vectors.

[0018] (2) The application significantly improves the efficiency and accuracy of crack detection, and is suitable for structural health monitoring fields such as bridges and buildings. BRIEF DESCRIPTION OF DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings used in the embodiments or prior art description will be briefly introduced. Obviously, the drawings in the following description are some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0020] Figure 1 A flowchart of a plane conversion method taking a crack plane as a reference coordinate system is provided for the embodiments of the present application.

[0021] Figure 2 A first sub-flowchart of a plane conversion method taking a crack plane as a reference coordinate system is provided for the embodiments of the present application.

[0022] Figure 3 A second sub-flowchart of a plane conversion method taking a crack plane as a reference coordinate system is provided for the embodiments of the present application.

[0023] Figure 4 A schematic diagram of a crack space coordinate system is provided for the embodiments of the present application.

[0024] Figure 5 A calculation process auxiliary line diagram is provided for the embodiments of the present application.

[0025] Figure 6 A schematic diagram of a camera coordinate system is provided for the embodiments of the present application. DETAILED DESCRIPTION

[0026] It should be noted that:

[0027] The terms "include" and "have" and any variations thereof in the specification and claims of the present application and the above drawings are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device including a series of steps or units does not have to include all the steps or units clearly listed, but can include other steps or units not clearly listed or inherent to the process, method, product or device.

[0028] The block diagrams shown in the drawings are merely functional entities, and do not necessarily correspond to physically independent entities. That is, the functional entities can be implemented in the form of software, or in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices. The flowcharts shown in the drawings are merely exemplary illustrations, and do not necessarily include all contents and operations / steps, nor are they necessarily executed in the order described. For example, some operations / steps can be further divided, and some operations / steps can be combined or partially combined, so that the actual execution order can be changed according to actual conditions.

[0029] In order to make the purposes, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present application. In addition, the technical features in each embodiment or in a single embodiment provided by the present application can be combined with each other at will to form new technical solutions, and such combination is not restricted by the order of steps and / or structure composition mode, but must be based on the fact that it can be realized by those skilled in the art. When the combination of technical solutions contradicts each other or cannot be realized, it should be considered that such combination of technical solutions does not exist, nor falls within the scope of protection required by the present application.

[0030] Please refer to the accompanying drawings Figure 1 The present application provides a plane conversion method taking a crack plane as a reference coordinate system (including steps S101-S105).

[0031] In step S101, the crack space coordinate system is determined through the camera pose when the first crack image is captured, and the space coordinates of each point in the crack space coordinate system are represented through the space relative relationship between the crack plane and the camera plane.

[0032] In step S101, the relative position of the camera and the crack plane when the first crack image is captured is determined as the y-axis; wherein the projection point of the camera on the crack plane when the first crack image is captured is the origin o; a parallel line of the horizontal frame of the camera frame passing through the origin on the crack plane is the x-axis; a vertical line of the x-axis passing through the origin on the crack plane is the z-axis, and a crack space coordinate system as shown in FIG. 1 is established. Figure 4 N in the crack space coordinate system represents the camera, and the coordinates of the camera N point are (0, 0, 0). A, B, E, F represent the four vertices (the four vertices are the positions of the four range finders) of the camera frame. The xoz plane is taken as the crack plane. A, B, E, F represent the four vertices (the four vertices are the positions of the four range finders) of the camera frame. The xoz plane is taken as the crack plane. For the angle between the angle bisector of the AOF and the xoy plane, respectively, A, B, E, F four vertexes on the infrared projection point on the crack plane. It should be noted that the crack plane represents the plane of the crack (i.e. the plane), the camera plane represents the plane of the camera and the camera frame (i.e. ABEF plane).

[0033] As shown in the accompanying Figure 5 , the projection points of the four vertices A, B, E, F on the xoy plane are respectively , the projection points of the four vertices A, B, E, F on the xoy plane are respectively is rotated to make and parallel to the x-axis after the corresponding four points. is half, is the angle between the angle bisector of the AOF and the y-axis. It can be understood that the angle between the camera plane and the crack plane is a three-dimensional angle, which is divided into three directions during the calculation process. The present application specifies that the upper and lower planes of the camera frame are parallel to the xoy plane during the establishment of the coordinate system, so there is no deflection with the y-axis as the center, the two angles respectively represent the deflection angles of the camera plane relative to the crack plane on the x-axis and z-axis. And is an angle set for convenience of calculation, without actual meaning.

[0034] Please refer to the accompanying Figure 2 , the specific steps of step S101 (including steps S1011-S1013) will be further introduced below.

[0035] Step S1011, the spatial coordinates of each point on the camera plane are expressed through the spatial relative relationship between the crack plane and the camera plane.

[0036] It should be noted that each point on the camera plane refers to all points on the camera plane, and these points have corresponding projection points on the crack plane. In this embodiment, the four vertices A, B, E, F on the camera frame are taken as an example for introduction, and the spatial coordinates of the four vertices A, B, E, F on the camera frame are expressed through the spatial relative relationship between the crack plane and the camera plane. The spatial relative relationship refers to the deflection angles of the camera plane relative to the crack plane on the x-axis and z-axis .

[0037] The spatial coordinates of point A are: ; ; . Among them, represents the coordinate of point A in the x-axis direction. represents the coordinate of point A in the y-axis direction. represents the coordinate of point A in the z-axis direction. L represents the side length of the camera frame. represents the coordinate of point N of the camera in the y-axis direction.

[0038] The spatial coordinates of point B are: ; ; . Wherein, represents the coordinate of point B in the x-axis direction. represents the coordinate of point B in the y-axis direction. represents the coordinate of point B in the z-axis direction.

[0039] The spatial coordinates of point E are: ; ; . Wherein, represents the coordinate of point E in the x-axis direction. represents the coordinate of point E in the y-axis direction. represents the coordinate of point E in the z-axis direction.

[0040] The spatial coordinates of point F are: ; ; . Wherein, represents the coordinate of point F in the x-axis direction. represents the coordinate of point F in the y-axis direction. represents the coordinate of point F in the z-axis direction.

[0041] Step S1012, calculate the plane normal vector of the camera plane, and based on the plane normal vector and the spatial coordinates of each point on the camera plane, respectively represent the straight line equation of the projection line of each point on the camera plane relative to the crack plane in the crack space coordinate system.

[0042] In this embodiment, the plane normal vector of the camera plane is calculated, and based on the plane normal vector and the spatial coordinates of the four vertices, the straight line equations of the infrared rays emitted by the range finder corresponding to the four vertices A, B, E, F on the camera frame in the crack space coordinate system are respectively represented. The plane normal vector of the camera plane is: . Given the plane normal vector and a point on the straight line, the straight line equation of the infrared ray emitted by the range finder at the position of the four vertices on the camera frame in the space is calculated.

[0043] The straight line equation of point A is: .

[0044] The straight line equation of point B is: .

[0045] The straight line equation of point E is: .

[0046] The equation of the straight line where point F is located is: .

[0047] In step S1013, the spatial coordinates of the projection points of each point on the camera plane on the crack plane are respectively represented based on the equation of the straight line.

[0048] In the embodiment, the spatial coordinates of the projection points of the four vertices on the camera frame on the crack plane are respectively represented based on the equation of the straight line. It can be understood that the infrared projection points of the four vertices A, B, E, and F on the camera frame on the crack plane xoz satisfy the four straight line equations respectively, and the four projection points are on the xoz plane (i.e., y is equal to 0). Therefore, by bringing y equal to 0 into the four straight line equations respectively, the coordinates of the projection points

[0049] Coordinates:

[0050] .

[0051] Coordinates:

[0052] .

[0053] Coordinates:

[0054] .

[0055] Coordinates:

[0056] .

[0057] In step S102, a camera coordinate system is established with one vertex on the camera frame as the origin, and the spatial relative relationship between the crack plane and the camera plane is calculated in the camera coordinate system.

[0058] In step S102, a camera coordinate system is established with one vertex on the camera frame as the origin, and the spatial relative relationship between the crack plane and the camera plane is calculated in the camera coordinate system. Figure 6 In step S102, a camera coordinate system is established with one vertex on the camera frame as the origin, and the spatial relative relationship between the crack plane and the camera plane is calculated in the camera coordinate system. The four points are the projection points of the range finders on the crack plane, ​​These represent the distances from the camera plane to the crack plane measured by the four rangefinders (i.e., the distances from the four vertices A, B, E, and F to the projection point). (The distance between them). Point for The projection point of the point onto the xoy plane, connect and point, That is Angle. Passing Draw a straight line parallel to the x-axis intersecting the point. The extension of the line at point, That is Angle. Spatial relative relationships include the angle of deflection of the camera plane relative to the crack plane on the x-axis. and the deflection angle on the z-axis .

[0059] Furthermore, The values ​​can be obtained from four distances by the rangefinder. The side length L of the crack identification instrument frame is obtained as follows: ; ,in, This represents the distance from the camera plane to the crack plane, as measured by the rangefinder at vertex A. This represents the distance from the camera plane to the crack plane as measured by the rangefinder at vertex B. This represents the distance from the camera plane to the crack plane as measured by the rangefinder at vertex E. This represents the distance from the camera plane to the crack plane as measured by the rangefinder at vertex F.

[0060] Step S103: Substitute the spatial relative relationship between the crack plane and the camera plane into the crack spatial coordinate system to obtain the spatial coordinates of each point in the crack spatial coordinate system.

[0061] In this embodiment, the spatial relative relationship between the crack plane and the camera plane, i.e., the deflection angle of the camera plane relative to the crack plane on the x-axis obtained in step S102, is defined. and the deflection angle on the z-axis Substituting the values ​​obtained in step S101 into the equations... The spatial coordinate expressions of the four projection points are used to calculate the spatial coordinates of the projection points corresponding to the four vertices on the camera frame in the crack spatial coordinate system, thereby realizing the transformation between the camera coordinate system and the crack spatial coordinate system.

[0062] Step S104: Establish a camera coordinate system for each crack image captured subsequently, calculate the absolute scale of each crack image, and perform correction processing on the crack image based on the absolute scale.

[0063] In step S104, a camera coordinate system is established with one of the top points of the camera external frame as the origin, as shown in the attached Figure 6 figure. The meanings of A, B, E, F, , , and L are the same as those of the camera coordinate system established in step S102, and will not be described here. The spatial coordinate information of each point in the established coordinate system is: A (L, 0, L), B (L, 0, 0), E (0, 0, 0), F (0, 0, L), (L, , L), (L, , 0), (0, , 0), (0, , L). It should be noted that whether it is the first crack image or each crack image subsequently photographed, the corresponding absolute scale needs to be calculated and the corresponding crack image needs to be corrected based on the absolute scale.

[0064] Please refer to the attached Figure 3 , the specific steps of calculating the absolute scale of each crack image and correcting the crack image based on the absolute scale (including steps S1041-S1046) will be further introduced below.

[0065] Step S1041, based on the four distances obtained by the camera frame side length and the range finder, the four edge lengths and four corners of the projection plane of the camera frame on the crack plane are calculated. Specifically, the angle of any corner of the camera frame projection plane is represented by the vector coordinates of the two adjacent edge lengths of the camera frame projection plane. . According to the parallelogram four-corner relationship, one angle is known, and the other three angles can be obtained. = , = = - . Finally, the coordinates of each point are used to obtain the four edge lengths. = , = , = , = .

[0066] Step S1042, based on the four edge lengths and four corners of the projection plane, the absolute scale of the crack image is obtained. It should be noted that the absolute scale of the crack image refers to the overall framework of the crack image. Obviously, given the four edge lengths and four corners, the absolute scale of the crack image can be obtained.

[0067] Step S1043, obtain the undistorted coordinates of each pixel point in the crack image. Specifically, first obtain the intrinsic parameters and distortion parameters of the camera, use a checkerboard calibration board, take multiple images of the calibration board at different angles, and use tools such as OpenCV to calculate the intrinsic matrix (including focal length and principal point coordinates) and distortion coefficients (radial distortion coefficients and tangential distortion coefficients) of the camera through a calibration algorithm. According to the intrinsic matrix and distortion parameters of the camera, a distortion model is established, and each pixel in the crack image is converted from the distortion coordinate system (i.e. the camera coordinate system) to the ideal undistorted coordinate system (i.e. the crack space coordinate system). For each pixel point in the crack image, the corresponding ideal undistorted coordinates are calculated according to the distortion model (i.e. the coordinates of each pixel point in the crack image in the crack space coordinate system). It can be understood that the coordinates of each pixel point in the first crack image in the crack space coordinate system are obtained through steps S101 to S103, which are mainly converted according to the relative relationship between the camera plane and the crack plane. The coordinates of each pixel point in each subsequent crack image are obtained through step S1043, which are converted according to the relationship between each subsequent crack image and the previous crack image. The coordinate acquisition methods of the first crack image and each subsequent crack image are different.

[0068] Step S1044, obtain the pixel value at the undistorted coordinates from the pre-corrected crack image, and assign it to the pixel point position in the post-corrected crack image through linear interpolation. Specifically, the pixel value at the undistorted coordinates is obtained from the original distorted image (i.e. the crack image before correction) through linear interpolation, and is assigned to the pixel point position in the corrected image.

[0069] Step S1045, establish the proportional relationship between image pixels and actual size based on absolute scale. Specifically, by measuring the pixel size of the object with known actual size (i.e. absolute scale) in the crack image, the scale (i.e. proportional relationship) between image pixels and actual size is calculated.

[0070] Step S1046, convert the coordinates of each pixel point in the crack image to actual physical coordinates based on the proportional relationship. Specifically, multiply the coordinates of the corrected crack image in the crack space coordinate system by the scale to convert them to coordinates in the actual physical coordinate system, thereby restoring the distortion deformation of the corrected crack image caused by the camera shooting center projection, and obtaining the real feature information of the crack.

[0071] Step S105, stitch multiple corrected crack images in the crack plane coordinate system through feature matching, and then convert them to the crack space coordinate system through the relative relationship between the stitched crack images.

[0072] In step S105, the multiple corrected crack images are stitched by feature matching in the crack plane coordinate system to obtain a complete crack image, and the complete crack image is converted into the crack space coordinate system by the relative relationship between the stitched crack images to obtain the complete image of the detected object (crack), complete crack features and distribution information. Understandably, a complete crack can be large, and the entire crack needs to be divided into multiple crack segments, and the images of the multiple crack segments are stitched together to obtain a complete crack image. The relative relationship refers to the relative positional relationship between different crack segments. The relative positional relationship between the crack images can be obtained according to the relationship between the coordinate system established according to the crack image and the coordinate system established according to the previous crack image and the stitching points of the feature matching of different crack images.

[0073] Based on the same inventive concept as the foregoing embodiments, the present application also provides a plane conversion device with a crack plane as a reference coordinate system. The plane conversion device includes a monocular crack recognition instrument and a terminal. The monocular crack recognition instrument is used to detect cracks and, when a crack is detected, capture a crack image and measure distance information from the crack plane. The terminal is used to obtain the crack image and the distance information from the monocular crack recognition instrument and execute a plane conversion method with a crack plane as a reference coordinate system.

[0074] Further, the monocular crack recognition instrument includes a monocular camera, four range finders, and a sensing structure. The monocular camera is fixed at the center of a square external frame (the camera frame described above refers to the external frame of the monocular crack recognition instrument) and is connected to four slide rails located on the diagonal lines of the external frame. The four range finders are arranged on the four slide rails. The sensing structure is used to detect cracks and, when a crack is detected, trigger the monocular camera to capture a crack image and trigger the four range finders to measure distance information from the crack plane. A handheld grip is provided on the external frame, and the sensing structure is configured on the handheld grip.

[0075] Based on the same inventive concept as the foregoing embodiments, the present application also provides a plane conversion system with a fracture plane as a reference coordinate system, the plane conversion system comprising a fracture space coordinate system establishing module, a camera coordinate system establishing module, a plane conversion module, a fracture image correction module, and a fracture image stitching module. The fracture space coordinate system establishing module is configured to determine a fracture space coordinate system through a camera pose when a first fracture image is captured, and to express space coordinates of each point in the fracture space coordinate system through a space relative relationship between a fracture plane and a camera plane. The camera coordinate system establishing module is configured to establish a camera coordinate system with a vertex on a camera frame as an origin, and to calculate the space relative relationship between the fracture plane and the camera plane in the camera coordinate system. The plane conversion module is configured to substitute the space relative relationship between the fracture plane and the camera plane into the fracture space coordinate system to obtain the space coordinates of each point in the fracture space coordinate system. The fracture image correction module is configured to establish a camera coordinate system for each fracture image captured subsequently, to calculate an absolute scale of each fracture image, and to perform correction processing on the fracture image based on the absolute scale. The fracture image stitching module is configured to stitch a plurality of corrected fracture images through feature matching in the fracture plane coordinate system, and to convert the stitched fracture images to the fracture space coordinate system through a relative relationship between the stitched fracture images.

[0076] Based on the same inventive concept as the foregoing embodiments, the present application also provides a non-transitory computer readable storage medium storing computer instructions, the computer instructions causing a computer to execute a plane conversion method with a fracture plane as a reference coordinate system.

[0077] In summary, the application discloses a plane conversion method taking a crack plane as a reference coordinate system, and realizes high-precision image stitching and three-dimensional reconstruction by establishing a crack space coordinate system and calculating infrared projection point coordinates and normal vectors. The application establishes a method for determining crack size by directly establishing an absolute scale based on an auxiliary structure. First, the camera pose when the first crack image is captured is used to determine the crack space coordinate system, and the coordinates of each point in the crack space coordinate system are expressed by the spatial relative relationship between the crack plane and the instrument plane. A camera coordinate system is established by taking one point of the camera external frame as the origin, and the spatial relative relationship between the crack plane and the instrument plane is calculated in the camera coordinate system. The spatial relative relationship between the crack plane and the instrument plane is substituted into the crack space coordinate system to obtain the spatial coordinates of each point in the crack space coordinate system. The absolute scale of the crack image is calculated by establishing a camera coordinate system for a single crack image, and the image is processed based on the absolute scale. The processed crack images are stitched by feature matching in the crack plane coordinate system, and the relative relationship between the stitched images is converted into the crack space coordinate system to obtain the complete image of the detected object and complete crack features and distribution information.

[0078] Finally, it should be noted that: the above examples are only used to illustrate the technical solutions of the application, and not to limit them; although the application has been described in detail with reference to the foregoing examples, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing examples, or make equivalent replacement for part or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solution deviate from the technical solutions of the embodiments of the application.

Claims

1. A plane conversion method of a fracture plane reference coordinate system, characterized by, The application relates to a method for generating a three-dimensional fracture image, comprising the following steps: determining a fracture space coordinate system through a camera posture when a first fracture image is captured, and representing the space coordinates of each point in the fracture space coordinate system through a space relative relationship between a fracture plane and a camera plane; establishing a camera coordinate system with a vertex on a camera frame as an origin point, and calculating the space relative relationship between the fracture plane and the camera plane in the camera coordinate system; substituting the space relative relationship between the fracture plane and the camera plane into the fracture space coordinate system to obtain the space coordinates of each point in the fracture space coordinate system; establishing a camera coordinate system for each fracture image captured subsequently, calculating the absolute scale of each fracture image, and performing correction processing on the fracture image based on the absolute scale; splicing the multiple corrected fracture images in a fracture plane coordinate system through feature matching, and converting the relative relationship between the spliced fracture images to the fracture space coordinate system.

2. The plane conversion method of claim 1, wherein The method for determining the fracture space coordinate system through the camera posture when the first fracture image is captured comprises the following steps: determining a y-axis according to the relative position between the camera and the fracture plane when the first fracture image is captured; wherein the projection point of the camera on the fracture plane is an origin point; a parallel line of a horizontal frame of the camera frame passing through the origin point on the fracture plane is an x-axis; and a vertical line of the x-axis passing through the origin point on the fracture plane is a z-axis, so as to establish the fracture space coordinate system.

3. A plane conversion method of a fracture plane reference coordinate system according to claim 2, wherein, The method for representing the space coordinates of each point in the fracture space coordinate system through the space relative relationship between the fracture plane and the camera plane comprises the following steps: representing the space coordinates of each point on the camera plane through the space relative relationship between the fracture plane and the camera plane; calculating a plane normal vector of the camera plane, and representing the straight line equation of the projection line of each point on the camera plane relative to the fracture plane in the fracture space coordinate system based on the plane normal vector and the space coordinates of each point on the camera plane; representing the space coordinates of the projection point of each point on the camera plane on the fracture plane based on the straight line equation.

4. The plane conversion method of claim 1, wherein The method for establishing the camera coordinate system with a vertex in the camera external frame as the origin point comprises the following steps: establishing the camera coordinate system with a vertex in the camera frame as the origin point, a parallel line of a horizontal frame of the camera frame passing through the origin point on the camera plane as the x-axis, a parallel line of a vertical frame of the camera frame passing through the origin point on the camera plane as the z-axis, and a line perpendicular to the camera plane and passing through the origin point as the y-axis.

5. The plane conversion method of claim 4, wherein The method for calculating the space relative relationship between the fracture plane and the camera plane in the camera coordinate system comprises the following steps: calculating the deflection angle of the camera plane relative to the fracture plane on the x-axis based on the side length of the camera frame and the four distances measured by the distance measuring instrument; calculating the deflection angle of the camera plane relative to the fracture plane on the z-axis based on the side length of the camera frame and the four distances measured by the distance measuring instrument; the space relative relationship comprises the deflection angles of the camera plane relative to the fracture plane on the x-axis and the z-axis.

6. The method of claim 1, wherein the method further comprises: The method for calculating the absolute scale of each fracture image comprises the following steps: calculating the four side lengths and four corners of the projection plane of the camera frame on the fracture plane based on the side length of the camera frame and the four distances measured by the distance measuring instrument; obtaining the absolute scale of the fracture image based on the four side lengths and four corners of the projection plane.

7. A plane conversion method of a fracture plane reference coordinate system according to claim 6, wherein, The crack image is corrected based on the absolute scale, including: Obtaining the non-distortion coordinates of each pixel point in the crack image; Obtaining the pixel value at the non-distortion coordinates from the crack image before correction, and assigning the pixel value to the pixel point position in the crack image after correction through linear interpolation; Establishing the proportional relationship between the image pixel and the actual size based on the absolute scale; Converting the coordinates of each pixel point in the crack image into actual physical coordinates based on the proportional relationship.

8. A plane conversion device of a coordinate system based on a fracture plane, characterized by Comprising: A monocular crack identification instrument for detecting cracks and, when a crack is detected, capturing a crack image and measuring distance information from the crack plane; A terminal for obtaining the crack image and distance information from the monocular crack identification instrument and executing a plane conversion method with the crack plane as the reference coordinate system according to any one of claims 1 to 7.

9. A plane conversion system with a fracture plane as a reference coordinate system, characterized by Comprising: A crack space coordinate system establishment module for determining the crack space coordinate system through the camera pose when the first crack image is captured, and expressing the space coordinates of each point in the crack space coordinate system through the space relative relationship between the crack plane and the camera plane; A camera coordinate system establishment module for establishing the camera coordinate system with one vertex on the camera frame as the origin, and calculating the space relative relationship between the crack plane and the camera plane in the camera coordinate system; A plane conversion module for substituting the space relative relationship between the crack plane and the camera plane into the crack space coordinate system to obtain the space coordinates of each point in the crack space coordinate system; A crack image correction module for establishing the camera coordinate system for each crack image captured subsequently, calculating the absolute scale of each crack image, and correcting the crack image based on the absolute scale; A crack image stitching module for stitching multiple crack images that have been corrected through feature matching in the crack plane coordinate system, and converting the stitched crack images to the crack space coordinate system through the relative relationship between the stitched crack images.

10. A non-transitory computer-readable storage medium, comprising: The non-transitory computer readable storage medium stores computer instructions, and the computer instructions cause the computer to execute the plane conversion method with the crack plane as the reference coordinate system according to any one of claims 1 to 7.

Images