A method for extracting temperature field of arch bridge based on infrared thermal imaging

Abstract

The present invention relates to the field of temperature field measurement, and specifically to a temperature field extraction method for an arch bridge based on infrared thermal imaging. The method uses a drone equipped with an infrared camera to collect local infrared images of the arch bridge, and at the same time, the automatic extraction of infrared image temperature data is realized through a DJI TSDK program, a PYTHON program, a MATLAB program, and a PHST software. The present invention performs ortho-transformation processing on the local temperature image, so that the local temperature image splicing is more consistent, and the overall temperature image of the arch bridge is better obtained. Finally, the temperature value of the design model coordinate is automatically extracted through a corresponding temperature extraction program.

Description

Technical Field

[0001] The invention relates to the technical field of temperature field measurement, and in particular to a method for extracting the temperature field of an arch bridge based on infrared thermal imaging. Background Art

[0002] Arch bridges have always played an important role in connecting cities and promoting communication in human civilization. The technological innovation of their design and construction has promoted the development of engineering disciplines. As one of the classics in architectural engineering, it continues to show its perfect combination of elegance and mechanics in the city skylines around the world. With the deepening of transportation construction, mountain bridges have become the norm. Arch bridges have become the vanguard because of the winding terrain of mountainous areas and have advantages that other bridge types do not have. The main characteristics of arch bridges include curved arch structures, uniform load distribution, excellent strength and stiffness, etc. These characteristics give arch bridges significant advantages when crossing complex terrains such as large rivers, canyons and mountain roads.

[0003] As the design span of bridges increases, the impact of temperature loads on bridge structures becomes more and more significant. Temperature changes can cause steel to expand or contract, which in turn causes the structure to expand and deform. If this temperature difference is ignored and only a uniform temperature model is used for finite element analysis, it may lead to errors with the actual stress conditions of the bridge. Traditional point temperature measurement methods, such as surface-mounted temperature sensors, have limitations such as low efficiency when measuring the temperature field of arch bridges. Summary of the invention

[0004] The purpose of an embodiment of the present invention is to provide a method for extracting the temperature field of an arch bridge based on infrared thermal imaging. By taking infrared images of the arch bridge, intensive measurement of its surface temperature field is achieved. The method includes data collection and image processing in four directions (up, down, left, and right) of the arch bridge, thereby establishing a method for extracting the temperature field of the arch bridge using infrared thermal imaging. At the same time, corresponding local temperature image recognition and image orthorectification programs are developed, so that the local infrared temperature images obtained during the shooting process can be better spliced, and an automatic extraction method for the temperature field is developed, thereby realizing rapid and accurate measurement and extraction of the temperature field of the arch bridge.

[0005] The embodiment of the present invention is achieved as follows:

[0006] An embodiment of the present invention provides a method for extracting the temperature field of an arch bridge based on infrared thermal imaging, comprising the following steps:

[0007] Step 1: Use a drone with an infrared lens to collect infrared images of the top, bottom, left and right sides of the bridge. The collection path should be taken from left to right in sequence. The shooting path should be parallel to the bridge body to ensure that the proportion of the arch bridge in each infrared image is consistent and that the overlap between two adjacent infrared images is 23%-26%;

[0008] Step 2: Use the DJI Thermal SDK program module to convert the captured R-JPEG infrared image into a RAW format image;

[0009] Step 3: Use the Python program module to convert the obtained RAW format image into a TIF format image, and process the converted TIF format image into a grayscale image with a grayscale value of 0 to 255;

[0010] Step 4: Using the MATLAB program module, perform orthographic projection processing on the grayscale image obtained in step 3;

[0011] Step 5: Use the PHST software module to stitch the local grayscale images after orthographic projection into an overall grayscale image;

[0012] Step 6: Use the MATLAB program module to perform image mapping on the stitched overall grayscale image, and restore the grayscale image to a temperature image;

[0013] Step 7: According to the coordinates of the arch bridge design model, the MATLAB program module is used to automatically extract the temperature values ​​of the corresponding coordinate points, thereby realizing automatic extraction of the temperature values.

[0014] Preferably, the method of performing orthographic projection processing on the obtained grayscale image in step 4 is as follows:

[0015] A1. Use the imread function to first read the local grayscale image image in the target directory, and input the calibrated camera intrinsic parameter matrix K = [a 11 ,a 12 ,a 13 ; a 21 ,a 22 ,a 23 ; 0,0,1] and the camera distortion parameters C = [b1,b2,b3,b4,b5], a in the internal parameter matrix K 11 and a 22 Represents the length of the camera focal length in the x-axis and y-axis directions, a 12 and a 21 represents the skew coefficient of the camera, a 13 and a 23 Indicates the coordinates of the principal point of the image, in the distortion parameter C, b1 and b2 represent the main radial distortion coefficients, b5 represents the secondary radial distortion coefficient, and b3 and b4 represent the tangential distortion coefficients;

[0016] A2. Get the focal length of the camera. Define FocalLenth = [K(1,1)K(2,2)] = [(a 11 )(a 22)] and the principal point definition PrincipalPoint = [K(1,3)K(2,3)] = [(a 13 )(a 23 )], extract the size of the local grayscale image and construct the camera's intrinsic parameter matrix object:

[0017] Intrinsics=cameraIntrinsics(FocalLength,PrincipalPoint,ImageSize);

[0018] A3. Construct the cameraParameters data structure of the camera:

[0019] Params = struct (Intrinsics, RP, TP), which includes the camera's intrinsic matrix Intrinsics, the camera's radial distortion parameter RP = B (b1: b2), and the camera's tangential distortion parameter TP = B (b3: b4);

[0020] A4. Use the undistortImage function in the MATLAB program module to correct the distortion of the original image and obtain a local grayscale image after distortion correction;

[0021] A5. Select 5 feature points along the axial direction of the arch bridge in the local grayscale image and obtain the pixel coordinates of the feature points PicPoints = [x1, y1; x2, y2; x3, y3; x4, y4; x5, y5] and the local world coordinates of the corresponding 5 feature points WorldPoints = [X1, 0, Z1; X2, 0, Z2; X3, 0, Z3; X4, 0, Z4; X5, 0, Z5];

[0022] A6. Use functions in MATLAB program modules

[0023] estworldpose(PicPoints,WorldPoints,Intrinsics) calculates the camera's posture and external parameters to obtain the camera rotation matrix CameraOR and translation vector CameraLoc;

[0024] A7. Use the function rotationMatrixToVector(CameraOR') in the MATLAB program module to calculate the relative rotation angle between the camera coordinate system and the world coordinate system to obtain the relative rotation angle rotationVector.

[0025] A8. Construct the rotation axis vectors rotationAxis about the x, y, and z axes respectively, and use the function vrrotvec2mat ([rotationAxis, rotationVector]) in the MATLAB program module to rotate the x, y, and z axes in turn, so as to obtain the world coordinates RotatedWorldPoints of the rotated local grayscale image;

[0026] A9. Reuse the functions in the MATLAB program module

[0027] estworldpose(PicPoints,RotatedWorldPoints,Intrinsics) calculates the camera's posture and external parameters to obtain a new camera rotation matrix NewCameraOR and a new translation vector NewCameraLoc;

[0028] A10. According to the new rotation matrix and translation vector, construct the transformation matrix WorldToCameraMatrix = [NewCameraOR', NewCameraLoc'; 0 0 0 1] from the world coordinate system to the camera coordinate system. Transform the point in the world coordinate system to the camera coordinate system through the transformation matrix, obtain the position of the transformed point in the camera coordinate system and save it to CameraPointsHomogeneous, extract the x and y axis coordinates in CameraPointsHomogeneous, and save them to TranspicPoints.

[0029] A11. Calculate the coordinate difference delta between each point in the image and the first feature point through TranspicPoints-TranspicPoints(1,:), and calculate the coordinate P of the transformed image feature point through delta+picpoints(1,:);

[0030] A12. Use functions in MATLAB program modules

[0031] estgeotform2d(picpoints,P,'projective'), calculate tform;

[0032] A13. Use the function imwarp(image,tform) in the MATLAB program module to process the orthogonal local grayscale image and realize the orthogonal processing of the local grayscale image.

[0033] Preferably, the method for performing image temperature mapping on the spliced ​​overall temperature image in step 6 is as follows:

[0034] B1. Using the MATLAB program module, obtain the pixel coordinates DesignCoordinates of several bridge design image feature points and the pixel coordinates ImageCoordinates of the corresponding overall temperature splicing image feature points;

[0035] B2. Use functions in MATLAB program modules

[0036] fitgeotrans(ImageCoordinates,DesignCoordinates,'affine'), calculates the affine transformation matrix T = [a, b, t x ;c,d,t y ; 0,0,1], the affine transformation form is

[0037] [X;Y]=[a,b;c,d][x;y]+[t x ;t y ];

[0038] B3. Use the imwarp function in the MATLAB program module to transform the temperature stitching image to the bridge design image;

[0039] B4. Use the imbinarize function in the MATLAB program module to perform binarization on the transformed image, replace all grayscale values ​​outside the outer contour of the bridge design image target body with 0, and replace all grayscale values ​​inside the outer contour with 1, and traverse the pixel coordinates of the area with grayscale value 1 in the image, map the grayscale value of the corresponding pixel coordinates in the spliced ​​overall temperature image to the area with grayscale value 1 in the design image, and complete the image temperature mapping and fusion.

[0040] Preferably, the method for automatically extracting the temperature value in step seven is as follows:

[0041] C1. Use the MATLAB program module to import the temperature image and coordinate table to be extracted, and display the temperature image;

[0042] C2. Select two feature points in the temperature image and import the design model coordinates corresponding to the feature points, so as to calculate the scaling factor and translation factor of the design model coordinates and the temperature image coordinates;

[0043] C3. Calculate the image coordinates corresponding to the design model coordinates according to the scaling factor and the translation factor, and mark all coordinate points in the table;

[0044] C4. Automatically extract the temperature value of the marked point in the temperature image, thereby realizing automatic temperature extraction.

[0045] Preferably, in step 1, when collecting images, two adjacent infrared images contain a 25% overlap.

[0046] The arch bridge temperature field extraction method based on infrared thermal imaging provided by the present invention has the following beneficial effects:

[0047] The present invention utilizes a drone equipped with an infrared camera to efficiently and quickly obtain local infrared images of the arch bridge, and automatically extracts the temperature values ​​of the design model coordinate points through the DJI TSDK program, PYTHON program, MATLAB program, and PHST software, thereby solving the problems of low temperature measurement efficiency and poor economy in the traditional arch bridge temperature measurement method when performing dense measurement of the arch rib plane temperature field. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 A flow chart of a method for extracting the temperature field of an arch bridge based on infrared thermal imaging provided in an embodiment of the present invention.

[0049] Figure 2 The PYTHON code written in the present invention is used to convert the obtained RAW format file into a TIF format image, and the TIF format image is converted into a grayscale image instance diagram with a grayscale value between 0 and 255.

[0050] Figure 3 The present invention uses the MATLAB program to perform orthographic projection processing on the obtained grayscale image.

[0051] Figure 4 This is an example diagram of an orthographic projection process performed on the obtained grayscale image using the MATLAB program written in the present invention.

[0052] Figure 5 This is an example diagram of using PHST software in the present invention to stitch local temperature images after orthographic projection into an overall temperature image.

[0053] Figure 6 The present invention is a flow chart of image temperature mapping using the MATLAB program for the spliced ​​overall temperature image.

[0054] Figure 7 This is an example diagram of image temperature mapping of the spliced ​​overall temperature image using the MATLAB program in the present invention.

[0055] Figure 8 The present invention uses the MATLAB program to automatically extract the temperature values ​​of the corresponding coordinate points according to the coordinates of the arch bridge design model, thereby realizing the automatic extraction flow chart of the temperature value.

[0056] Fig. 9This is an example diagram of automatically extracting the temperature values ​​of corresponding coordinate points according to the coordinates of the arch bridge design model in the present invention using the MATLAB program, thereby realizing automatic extraction of temperature values. DETAILED DESCRIPTION

[0057] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0058] See also Figures 1 to 9 The embodiment of the present invention provides a method for extracting the temperature field of an arch bridge based on infrared thermal imaging, comprising the following steps:

[0059] Step 1: Use a drone with an infrared lens to collect infrared images of the top, bottom, left and right sides of the bridge. The collection path should be taken from left to right in sequence. The shooting path should be parallel to the bridge body to ensure that the proportion of the arch bridge in each infrared image is consistent and that the two adjacent infrared images have a 23%-26% overlap.

[0060] In this step, when collecting images, it is preferred that two adjacent infrared images contain a 25% overlap.

[0061] Step 2: Use the DJI Thermal SDK program module to convert the captured R-JPEG infrared image into a RAW format image.

[0062] Step 3: Use the PYTHON program module to convert the obtained RAW format image into a TIF format image, and process the converted TIF format image into a grayscale image with a grayscale value of 0 to 255.

[0063] Step 4: Use the MATLAB program module to perform orthographic projection processing on the grayscale image obtained in step 3.

[0064] In this step, the specific orthographic projection processing method is as follows:

[0065] A1. Use the imread function to first read the local grayscale image image in the target directory, and input the calibrated camera intrinsic parameter matrix K = [a 11 ,a 12 ,a 13 ; a 21 ,a 22 ,a 23; 0,0,1] and the camera distortion parameters C = [b1,b2,b3,b4,b5], a in the internal parameter matrix K 11 and a 22 Represents the length of the camera focal length in the x-axis and y-axis directions, a 12 and a 21 Indicates the camera's skew coefficient, usually set to 0, a 13 and a 23 Represents the coordinates of the principal point of the image, which is usually located at the center of the image, because the x-axis and y-axis of the image are usually orthogonal. In the distortion parameter C, b1 and b2 represent the main radial distortion coefficients, b5 represents the secondary radial distortion coefficient, and b3 and b4 represent the tangential distortion coefficients;

[0066] A2. Get the focal length of the camera. Define FocalLenth = [K(1,1)K(2,2)] = [(a 11 )(a 22 )] and the principal point definition PrincipalPoint = [K(1,3)K(2,3)] = [(a 13 )(a 23 )], extract the size of the local grayscale image and construct the camera's intrinsic parameter matrix object:

[0067] Intrinsics=cameraIntrinsics(FocalLength,PrincipalPoint,ImageSize);

[0068] A3. Construct the cameraParameters data structure of the camera:

[0069] Params = struct (Intrinsics, RP, TP), which includes the camera's intrinsic matrix Intrinsics, the camera's radial distortion parameter RP = B (b1: b2), and the camera's tangential distortion parameter TP = B (b3: b4);

[0070] A4. Use the undistortImage function in the MATLAB program module to correct the distortion of the original image and obtain a local grayscale image after distortion correction;

[0071] A5. Select 5 feature points along the axial direction of the arch bridge in the local grayscale image and obtain the pixel coordinates of the feature points PicPoints = [x1, y1; x2, y2; x3, y3; x4, y4; x5, y5] and the local world coordinates of the corresponding 5 feature points WorldPoints = [X1, 0, Z1; X2, 0, Z2; X3, 0, Z3; X4, 0, Z4; X5, 0, Z5];

[0072] A6. Use functions in MATLAB program modules

[0073] estworldpose(PicPoints,WorldPoints,Intrinsics) calculates the camera's posture and external parameters to obtain the camera rotation matrix CameraOR and translation vector CameraLoc;

[0074] A7. Use the function rotationMatrixToVector(CameraOR') in the MATLAB program module to calculate the relative rotation angle between the camera coordinate system and the world coordinate system to obtain the relative rotation angle rotationVector.

[0075] A8. Construct the rotation axis vectors rotationAxis about the x, y, and z axes respectively, and use the function vrrotvec2mat ([rotationAxis, rotationVector]) in the MATLAB program module to rotate the x, y, and z axes in turn, so as to obtain the world coordinates RotatedWorldPoints of the rotated local grayscale image;

[0076] A9. Reuse the functions in the MATLAB program module

[0077] estworldpose(PicPoints,RotatedWorldPoints,Intrinsics) calculates the camera's posture and external parameters to obtain a new camera rotation matrix NewCameraOR and a new translation vector NewCameraLoc;

[0078] A10. According to the new rotation matrix and translation vector, construct the transformation matrix WorldToCameraMatrix = [NewCameraOR', NewCameraLoc'; 0 0 0 1] from the world coordinate system to the camera coordinate system. Transform the point in the world coordinate system to the camera coordinate system through the transformation matrix, obtain the position of the transformed point in the camera coordinate system and save it to CameraPointsHomogeneous, extract the x and y axis coordinates in CameraPointsHomogeneous, and save them to TranspicPoints.

[0079] A11. Calculate the coordinate difference delta between each point in the image and the first feature point through TranspicPoints-TranspicPoints(1,:), and calculate the coordinate P of the transformed image feature point through delta+picpoints(1,:);

[0080] A12. Use functions in MATLAB program modules

[0081] estgeotform2d(picpoints,P,'projective'), calculate tform;

[0082] A13. Use the function imwarp(image,tform) in the MATLAB program module to process the orthogonal local grayscale image and realize the orthogonal processing of the local grayscale image.

[0083] Step 5: Use the PHST software module to stitch the local grayscale images after orthographic projection into an overall grayscale image.

[0084] Step 6: Use the MATLAB program module to perform image mapping on the stitched overall grayscale image and restore the grayscale image to a temperature image.

[0085] In this step, the specific processing method is as follows:

[0086] B1. Using the MATLAB program module, obtain the pixel coordinates DesignCoordinates of several bridge design image feature points and the pixel coordinates ImageCoordinates of the corresponding overall temperature splicing image feature points;

[0087] B2. Use functions in MATLAB program modules

[0088] fitgeotrans(ImageCoordinates,DesignCoordinates,'affine'), calculates the affine transformation matrix T = [a, b, t x ;c,d,t y ; 0,0,1], the affine transformation form is

[0089] [X;Y]=[a,b;c,d][x;y]+[t x ;t y ];

[0090] B3. Use the imwarp function in the MATLAB program module to transform the temperature stitching image to the bridge design image;

[0091] B4. Use the imbinarize function in the MATLAB program module to perform binarization on the transformed image, replace all grayscale values ​​outside the outer contour of the bridge design image target body with 0, and replace all grayscale values ​​inside the outer contour with 1, and traverse the pixel coordinates of the area with grayscale value 1 in the image, map the grayscale value of the corresponding pixel coordinates in the spliced ​​overall temperature image to the area with grayscale value 1 in the design image, and complete the image temperature mapping and fusion.

[0092] Step 7: According to the coordinates of the arch bridge design model, the MATLAB program module is used to automatically extract the temperature values ​​of the corresponding coordinate points, thereby realizing automatic extraction of the temperature values.

[0093] In this step, the method for automatically extracting the temperature value is as follows:

[0094] C1. Use the MATLAB program module to import the temperature image and coordinate table to be extracted, and display the temperature image;

[0095] C2. Select two feature points in the temperature image and import the design model coordinates corresponding to the feature points, so as to calculate the scaling factor and translation factor of the design model coordinates and the temperature image coordinates;

[0096] C3. Calculate the image coordinates corresponding to the design model coordinates according to the scaling factor and the translation factor, and mark all coordinate points in the table;

[0097] C4. Automatically extract the temperature value of the marked point in the temperature image, thereby realizing automatic temperature extraction.

[0098] In summary, by using the method proposed in the present invention, a drone equipped with an infrared camera can efficiently and quickly obtain local infrared images of the arch bridge. Through the DJI TSDK program, PYTHON program, MATLAB program, and PHST software, the temperature values ​​of the temperature measurement grid pixels can be automatically extracted, which solves the problems of low temperature measurement efficiency and poor economy in the traditional arch bridge temperature measurement method when performing dense measurement of the arch rib plane temperature field.

[0099] It should be noted that, in this article, the terms "comprises", "includes" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or apparatus that includes a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or apparatus.

[0100] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for extracting the temperature field of an arch bridge based on infrared thermal imaging, characterized in that: The following steps are involved: Step 1: Use a drone with an infrared lens to collect infrared images of the top, bottom, left and right sides of the bridge. The collection path should be taken from left to right in sequence. The shooting path should be parallel to the bridge body to ensure that the proportion of the arch bridge in each infrared image is consistent and that the overlap between two adjacent infrared images is 23%-26%; Step 2: Use the DJI Thermal SDK program module to convert the captured R-JPEG infrared image into a RAW format image; Step 3: Use the Python program module to convert the obtained RAW format image into a TIF format image, and process the converted TIF format image into a grayscale image with a grayscale value of 0 to 255; Step 4: Using the MATLAB program module, perform orthographic projection processing on the grayscale image obtained in step 3; Step 5: Use the PHST software module to stitch the local grayscale images after orthographic projection into an overall grayscale image; Step 6: Use the MATLAB program module to perform image mapping on the stitched overall grayscale image, and restore the grayscale image to a temperature image; Step 7: According to the coordinates of the arch bridge design model, the temperature values ​​of the corresponding coordinate points are automatically extracted using the MATLAB program module, thereby realizing automatic extraction of the temperature values; The method for performing orthographic projection processing on the obtained grayscale image in step 4 is as follows: A1. Use the imread function to first read the local grayscale image image in the target directory, and input the calibrated camera intrinsic parameter matrix K = [a 11 ,a 12 ,a 13 ; a 21 ,a 22 ,a 23 ; 0,0,1] and the camera distortion parameters C = [b1,b2,b3,b4,b5], a in the internal parameter matrix K 11 and a 22 Represents the length of the camera focal length in the x-axis and y-axis directions, a 12 and a 21 represents the camera's skew coefficient, a 13 and a 23 Indicates the coordinates of the principal point of the image, in the distortion parameter C, b1 and b2 represent the main radial distortion coefficients, b5 represents the secondary radial distortion coefficient, and b3 and b4 represent the tangential distortion coefficients; A2. Get the focal length of the camera. Define FocalLenth = [K(1,1)K(2,2)] = [(a 11 )(a 22 )] and the principal point definition PrincipalPoint = [K(1,3)K(2,3)] = [(a 13 )(a 23 )], extract the size of the local grayscale image and construct the camera's intrinsic parameter matrix object: Intrinsics=cameraIntrinsics(FocalLength,PrincipalPoint,ImageSize); A3. Construct the cameraParameters data structure of the camera: Params = struct (Intrinsics, RP, TP), which includes the camera's intrinsic matrix Intrinsics, the camera's radial distortion parameter RP = B (b1: b2), and the camera's tangential distortion parameter TP = B (b3: b4); A4. Use the undistortImage function in the MATLAB program module to correct the distortion of the original image and obtain a local grayscale image after distortion correction; A5. Select 5 feature points along the axial direction of the arch bridge in the local grayscale image and obtain the pixel coordinates of the feature points PicPoints = [x1, y1; x2, y2; x3, y3; x4, y4; x5, y5] and the local world coordinates of the corresponding 5 feature points WorldPoints = [X1, 0, Z1; X2, 0, Z2; X3, 0, Z3; X4, 0, Z4; X5, 0, Z5]; A6. Use functions in MATLAB program modules estworldpose(PicPoints,WorldPoints,Intrinsics) calculates the camera's posture and external parameters to obtain the camera rotation matrix CameraOR and translation vector CameraLoc; A7. Use the function rotationMatrixToVector(CameraOR') in the MATLAB program module to calculate the relative rotation angle between the camera coordinate system and the world coordinate system to obtain the relative rotation angle rotationVector. A8. Construct the rotation axis vectors rotationAxis about the x, y, and z axes respectively, and use the function vrrotvec2mat ([rotationAxis, rotationVector]) in the MATLAB program module to rotate the x, y, and z axes in turn, so as to obtain the world coordinates RotatedWorldPoints of the rotated local grayscale image; A9. Reuse the functions in the MATLAB program module estworldpose(PicPoints,RotatedWorldPoints,Intrinsics) calculates the camera's posture and external parameters to obtain a new camera rotation matrix NewCameraOR and a new translation vector NewCameraLoc; A10. According to the new rotation matrix and translation vector, construct the transformation matrix WorldToCameraMatrix = [NewCameraOR', NewCameraLoc'; 0 0 0 1] from the world coordinate system to the camera coordinate system. Transform the point in the world coordinate system to the camera coordinate system through the transformation matrix, obtain the position of the transformed point in the camera coordinate system and save it to CameraPointsHomogeneous, extract the x and y axis coordinates in CameraPointsHomogeneous, and save them to TranspicPoints. A11. Calculate the coordinate difference delta between each point in the image and the first feature point through TranspicPoints-TranspicPoints(1,:), and calculate the coordinate P of the transformed image feature point through delta+picpoints(1,:); A12. Use functions in MATLAB program modules estgeotform2d(picpoints,P,'projective'), calculate tform; A13. Use the function imwarp(image,tform) in the MATLAB program module to process the orthogonal local grayscale image and realize the orthogonal processing of the local grayscale image.

2. The arch bridge temperature field extraction method based on infrared thermal imaging according to claim 1 is characterized in that: The method for performing image temperature mapping on the stitched overall temperature image in step 6 is as follows: B1. Using the MATLAB program module, obtain the pixel coordinates DesignCoordinates of several bridge design image feature points and the pixel coordinates ImageCoordinates of the corresponding overall temperature splicing image feature points; B2. Use functions in MATLAB program modules fitgeotrans(ImageCoordinates,DesignCoordinates,'affine'), calculates the affine transformation matrix T = [a, b, t x ;c,d,t y ; 0,0,1], the affine transformation form is [X;Y]=[a,b;c,d][x;y]+[t x ;t y ]; B3. Use the imwarp function in the MATLAB program module to transform the temperature stitching image to the bridge design image; B4. Use the imbinarize function in the MATLAB program module to perform binarization on the transformed image, replace all grayscale values ​​outside the outer contour of the bridge design image target body with 0, and replace all grayscale values ​​inside the outer contour with 1, and traverse the pixel coordinates of the area with grayscale value 1 in the image, map the grayscale value of the corresponding pixel coordinates in the spliced ​​overall temperature image to the area with grayscale value 1 in the design image, and complete the image temperature mapping and fusion.

3. The arch bridge temperature field extraction method based on infrared thermal imaging according to claim 1 is characterized in that: The method for automatically extracting the temperature value in step 7 is as follows: C1. Use the MATLAB program module to import the temperature image and coordinate table to be extracted, and display the temperature image; C2. Select two feature points in the temperature image and import the design model coordinates corresponding to the feature points, so as to calculate the scaling factor and translation factor of the design model coordinates and the temperature image coordinates; C3. Calculate the image coordinates corresponding to the design model coordinates according to the scaling factor and the translation factor, and mark all coordinate points in the table; C4. Automatically extract the temperature value of the marked point in the temperature image, thereby realizing automatic temperature extraction.

4. The arch bridge temperature field extraction method based on infrared thermal imaging according to claim 1 is characterized in that: In step 1, when collecting images, two adjacent infrared images contain 25% overlap.

Images