An apparatus and method for geometric calibration of laboratory temperature and water volume infrared remote sensing measurement.

Abstract

The present application relates to a kind of laboratory warm drainage infrared remote sensing measurement geometric calibration device and method, comprising: warm drainage experimental water tank, infrared camera, thermal resistance geometric calibration point, the setting position of each thermal resistance geometric calibration point is: along water tank left and right sides at least one column thermal resistance geometric calibration point is distributed;Thermal resistance geometric calibration point along water tank long direction The spacing needs to be guaranteed in the infrared camera frame upper and lower direction at least can accommodate two rows of thermal resistance geometric calibration point;Thermal resistance geometric calibration point size is: at least 1 pixel in the frame of infrared camera.The present application uses thermal resistance in water body, by adjusting current, makes thermal resistance show obvious mark as geometric calibration point in infrared photo, to obtain accurate calibration on infrared photo, using these accurate calibration points to carry out a series of processing to the infrared photo taken, through these processing, high-quality warm drainage thermal field distribution diagram is obtained, the precision of warm drainage experiment is greatly improved.

Description

Technical Field

[0001] This invention relates to an apparatus and method for geometric calibration of infrared remote sensing measurement of laboratory thermal drainage, an apparatus and method for hydraulic experiments, and a measuring apparatus and method for simulating thermal drainage from power plants in a laboratory. Background Technology

[0002] Laboratory-scale thermal drainage model testing is one of the main methods for studying thermal drainage in power plants. The model water surface area typically exceeds 10m in width and 30m in length. The simplest method for measuring the temperature of the thermal drainage diffusion area in the model is to directly use multiple conventional contact temperature sensors to measure the temperature at multiple points in the model water. However, due to the limited number of measurement points, the accuracy of capturing the morphology of the thermal drainage is lacking. Infrared imaging can better identify the morphology of thermal drainage, compensating for the shortcomings of point measurements, but generally requires calibration through a small number of simultaneous temperature measurements at the water surface to ensure measurement accuracy. Existing laboratory infrared cameras mainly use non-orthophoto imaging; due to the lack of geometric correction, they cannot be combined with point measurements for quantitative determination of the influence range of thermal drainage and are only used as auxiliary observations. Infrared camera images identify objects through temperature; therefore, unlike visible light cameras, geometric calibration points cannot be set based on object color differences. Temperature difference is required to identify geometric calibration points. Existing infrared camera geometric calibration methods require specially designed infrared calibration boards and are generally only suitable for small shooting ranges. This method suffers from calibration deficiencies, resulting in low calibration accuracy in large-scale shooting, which is difficult to compensate for algorithmically. Furthermore, custom-made large infrared calibration boards are too costly and impractical. Therefore, achieving accurate geometric calibration and quantitative infrared thermometry in large scenes at a lower cost is a problem that needs to be solved. Summary of the Invention

[0003] To overcome the shortcomings of existing technologies, this invention proposes a device and method for geometric calibration of laboratory-grade thermal drainage infrared remote sensing measurements. The device and method combine infrared temperature field imaging with surface temperature measurements for precise temperature field identification, and utilize accurate calibration data to perform multiple image processing techniques, thereby obtaining a more accurate thermal drainage temperature field.

[0004] The objective of this invention is achieved as follows: a device for geometric calibration of laboratory temperature-dissipated water using infrared remote sensing, comprising: a temperature-dissipated water tank; an infrared camera capable of moving and capturing images along the length of the water tank positioned directly above the water tank; multiple resistance temperature detectors (RTDs) geometric calibration points connected to an adjustable power supply positioned at a certain height above the water surface in the water tank; each RTD geometric calibration point being positioned such that at least one row of RTDs geometric calibration points is distributed on each of the left and right sides along the width of the water tank; the spacing between the RTD geometric calibration points along the length of the water tank must ensure that at least two rows of RTDs geometric calibration points can be accommodated vertically within the frame of the infrared camera; and the geometric dimensions of each RTD geometric calibration point must be at least one pixel within the frame of the infrared camera.

[0005] Furthermore, the thermal resistors at each of the aforementioned thermal resistor geometric calibration points are connected to an adjustable power supply via series, parallel, or grouped series-parallel connections.

[0006] Furthermore, the thermal resistance geometric calibration point is supported by a bracket fixed to the bottom of the temperature drainage test tank.

[0007] Furthermore, the geometric calibration points of the thermal resistance are supported by ropes or wires fixed on both sides of the water tank.

[0008] A method for geometric calibration of laboratory temperature and water volume infrared remote sensing measurements using the above-mentioned device, the method comprising the following steps:

[0009] Step 1, set up the geometric calibration points of the thermal resistor: calculate the size of the infrared camera's shooting frame and the overlapping area by shooting height, design the arrangement position of the geometric calibration points of the thermal resistor according to the shooting frame, so that there are marker points near the four corners of each frame, and set up and record the coordinate position of the thermal resistor center point of the geometric calibration points of the thermal resistor.

[0010] Step 2, Adjust the geometric calibration point of the RTD: Adjust the RTD current through the geometric calibration point of the RTD so that the geometric calibration point of the RTD can be clearly displayed in the field of view of the infrared camera;

[0011] Step 3, conduct model tests: Flow water through a warm drainage experimental tank to simulate the diffusion pattern of warm drainage under environmental flow dilution, and use an infrared camera to capture the temperature field in the warm drainage experimental tank frame by frame along the length of the tank.

[0012] Step 4, Perform infrared image geometric calibration: This includes the following sub-steps:

[0013] Sub-step 1, Preparation of geometric coordinate data for water surface: Determine the geometric coordinate array of each image within the shooting area based on the coordinates of the water surface recording points and the actual shooting range;

[0014] Sub-step 2, Image preprocessing: Perform distortion correction and orientation normalization on the camera. By using the calibrated camera parameters, perform non-linear distortion correction on the original image and retain the image within the full field of view;

[0015] 2.1 Distortion correction:

[0016] Perform non-linear distortion correction on the original image by using the calibrated camera parameters;

[0017] 2.2 Orientation normalization:

[0018] Rotate the image according to the distribution direction of the landmark coordinates;

[0019] Sub-step 3, Geometric landmark identification: Combine automatic detection and manual detection. Automatically detect the geometric center of the circular object in the image as the image coordinates of the landmark. For the points that cannot be automatically identified, manually select points to complete;

[0020] Automatic detection:

[0021] Adopt an improved circular detection algorithm:

[0022] 3.1 Divide the matrix into a binary matrix according to a specific value. By adjusting the segmentation threshold, eliminate all water temperature change intervals so that the image only shows the graphic objects in the same interval as the temperature value and the thermoresistance temperature of the landmark;

[0023] 3.2 Delete all objects containing fewer than a specific number of pixels, that is, noise points. The set value of the specific number of pixels is determined by debugging during the initial processing;

[0024] 3.3 Fill the hole pixels inside the object by using object boundary tracking and region feature analysis and adopting morphological closing operation;

[0025] 3.4 Calculate the area of the object and screen for circles based on the circularity index of the ratio of the area A to the square of the perimeter P (M = 4πA / P 2 ) to identify circles. Adjust the value of M so that all landmarks can be identified;

[0026] 3.5 Set a low area threshold Amin and a high threshold Amax, and accurately screen out the landmark pixel cluster objects through double filtering of the object area A (Amin < A < Amax);

[0027] Artificial intervention mechanism:

[0028] When the automatic detection of some landmarks fails:

[0029] 3.6 Display candidate points on the RGB image;

[0030] 3.7 Manually complete the landmarks;

[0031] 3.8 Perform image center calculation on all calibration points and automatically sort and optimize the center coordinates so that the points are arranged according to a specific starting point and rotation direction;

[0032] Sub-step 4, Image Geometric Correction: This is divided into two stages. In the first stage, based on the ratio of the horizontal spacing dx0 to the vertical spacing dy0 represented by the geometric coordinates of the four marker points within the image frame, and the horizontal pixel coordinate range [x1, x2] and vertical pixel coordinate range [y1, y2] represented by the image coordinates of the four marker points, new marker point image coordinates are determined. This ensures that the horizontal and vertical spacing between marker points is not less than the distance between x1 and x2 and the distance between y1 and y2, respectively, and the ratio of the horizontal and vertical distances between marker points is close to dx0 / dy0. Based on the new marker point image coordinates, the geometrically corrected image is bilinearly interpolated and projected onto the new image. The purpose of this step is mainly to correct geometric distortions without significantly changing the image resolution. The second stage involves reprojecting the new image corrected in the first stage according to the geometric coordinates of the four marker points, transforming the image into the result of actual geometric coordinates, thus preparing for the stitching of multiple images.

[0033] Sub-step 5, Image Output: Generate an image file from the image with actual geometric coordinates to provide a basis for subsequent temperature correction between different amplitudes; at the same time, for the convenience of subsequent image stitching, based on the geometric coordinates of the 4 marker points, crop out the image within the quadrilateral area enclosed by the marker points, and output the cropped image as the image stitching output data after geometric correction.

[0034] 5.1 Full-frame output:

[0035] Generate an uncropped image file containing combined geometric coordinate information;

[0036] 5.2 Intelligent cropping:

[0037] Automatically search for the boundary range of the effective rectangular area represented by the calibration point, perform image cropping accordingly, and output cropped image file;

[0038] Sub-step 6, image stitching: After processing multiple images sequentially in the image processing software, the stitched image result is directly output by cropping and stitching the images.

[0039] The advantages and beneficial effects of this invention are as follows: This invention utilizes thermal resistors placed at appropriate positions on the water surface. By adjusting the current, the thermal resistors are made to appear as clear marks in infrared photographs as geometric calibration points, thereby obtaining precise calibration points on the infrared photographs. These precise calibrations are then used to perform a series of processing steps on the captured infrared photographs, including:

[0040] 1. Two-stage correction mechanism: the first projection eliminates geometric distortion, and the second transformation optimizes spatial accuracy;

[0041] 2. Hybrid detection algorithm: Composite detection of adaptive threshold segmentation + morphological optimization + circularity criterion;

[0042] 3. Intelligent orientation adaptation: Automatically adjusts the image orientation based on coordinate distribution;

[0043] 4. Dynamic resolution calculation: Real-time resolution calibration based on measured marker spacing;

[0044] 5. Lossless-Trimmed Dual Output: Meets the data needs of different application scenarios.

[0045] These processes yielded high-quality thermal field distribution maps for the warm drainage, significantly improving the accuracy of the warm drainage experiment. Since resistance temperature detectors (RTDs) are very inexpensive, their setup and installation costs are extremely low, greatly reducing the overall experimental cost and making it a highly cost-effective solution. Attached Figure Description

[0046] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0047] Figure 1 This is a structural schematic diagram of the device described in Embodiment 1 of the present invention, and a plan view of the water tank;

[0048] Figure 2 This is a structural schematic diagram of the device described in Embodiment 1 of the present invention, and an elevation view of the water tank;

[0049] Figure 3 This is a schematic diagram of the device described in Embodiment 1 of the present invention. Figure 1 The enlarged image at point A, rotated 90 degrees clockwise, is also a frame that the infrared camera could capture at point A;

[0050] Figure 4 This is a schematic diagram of the series connection circuit of the thermal resistors according to Embodiment 2 of the present invention;

[0051] Figure 5 This is a schematic diagram of the parallel connection circuit of the thermal resistors at the geometric calibration points of the thermal resistors described in Embodiment 2 of the present invention;

[0052] Figure 6 This is a schematic diagram of the circuit diagram for the thermal resistance grouping series and parallel connection of the thermal resistance geometric calibration points as described in Embodiment 2 of the present invention;

[0053] Figure 7 This is a schematic diagram of an original photograph taken by an infrared camera, showing lens distortion.

[0054] Figure 8 This is a schematic diagram of a photograph after distortion correction using the method described in Embodiment 5 of the present invention;

[0055] Figure 9This is a flowchart of the method described in Embodiment 5 of the present invention. Detailed Implementation

[0056] Example 1:

[0057] This embodiment is a device for geometric calibration of laboratory temperature and water displacement infrared remote sensing measurement, such as... Figure 1 , 2 The embodiment includes: a temperature-dissipation experimental water tank 1; an infrared camera 2, capable of moving and capturing images along the length of the water tank, positioned directly above the water tank; and multiple thermal resistance geometric calibration points 3, connected to an adjustable power supply, set at a certain height above the water surface in the water tank. The thermal resistance geometric calibration points are positioned such that at least one row of calibration points is distributed on each of the left and right sides along the width of the water tank; the spacing between the thermal resistance geometric calibration points along the length of the water tank must ensure that at least two rows of calibration points can be accommodated vertically within the frame of the infrared camera; and the geometric dimensions of each thermal resistance geometric calibration point must be at least one pixel within the frame of the infrared camera.

[0058] The water tank for temperature drainage experiments is usually a rectangular strip tank with flowing water inside. Figure 1 , 2 The hollow arrows in the center indicate the direction of water flow, used to simulate the ambient water flow of warm drainage. A simulated warm drainage outlet 4 is located on one side of the tank (e.g.,...). Figure 1 , 2 (As shown), this is a drainage channel or pipe outlet used to simulate warm water drainage. It should be noted that the front, back, left, and right directions described in this embodiment and the following embodiments are determined according to the direction of water flow in the tank.

[0059] A movable feature is installed above the water tank, capable of moving along the length of the water tank. Figure 1 , 2 An infrared camera (in the direction of the solid arrow) is used to image the temperature field in the water tank. Ideally, the infrared camera should be mounted on a bracket that slides along a track, with the track fixed directly above the water tank. During shooting, the bracket moves along the track segment by segment, allowing the captured images to be stitched together to form a complete vertical image of the water tank, providing excellent conditions for future image stitching. When the infrared camera's field of view cannot cover the width of the water tank, the camera needs to be moved laterally, and the images need to be stitched laterally during the stitching process to form a complete horizontal image of the water tank.

[0060] The key to this embodiment lies in setting up multiple thermal resistors (RTRs) above the water surface (generally 5-10 cm) as geometric calibration points. The RRTs are arranged in a vertical and horizontal pattern, with columns along the water flow direction and horizontal rows perpendicular to it. There are usually two columns, one on each side of the water tank, evenly distributed along the sidewalls. The columns can be symmetrically distributed or offset to the left or right as needed for the experiment. Figure 2 The left deviation is shown. Multiple rows are evenly distributed horizontally along the length of the tank, as shown... Figure 1 , 2 This displays an array of 16 horizontal rows of thermistors. The spacing between the two rows of thermistors is such that they are covered vertically by the infrared camera's image frame, and they are located near the top and bottom edges of the frame. In other words, at least four thermistors (two rows, two per row) should be visible in a single image captured by the infrared camera. These four thermistors should occupy the four corners of the image frame. If the entire image frame is divided into four parts, then each of the four thermistors occupies one part. Figure 3 As shown. These four RTDs form a set of RTD geometric calibration points, and the center coordinates of each set of RTD geometric calibration points are the intersection of the two diagonals.

[0061] The aforementioned thermal resistors generate different amounts of heat when different currents pass through them, and this heat generation is reflected differently in the infrared camera's photographs, appearing in different colors. Based on this characteristic, this embodiment connects the thermal resistor array together via wires. The connection method can be all in series, all in parallel, or, depending on the location of the thermal resistors, they can be connected in series in different regions and then in parallel, with different voltages and currents supplied to each region to form region identification. Because the infrared camera's frame and the geometric calibration point have a very definite positional relationship, the thermal resistors, which can serve as geometric calibration points, can generate very precise positional coordinates on the infrared photographs, thereby obtaining very accurate temperature field values ​​for the thermal discharge. Using these precise temperature field values, and through various processing methods on the captured infrared photographs, the accuracy of the thermal discharge experiment is greatly improved.

[0062] An adjustable power supply is a special power supply for laboratory use, whose output current can be steplessly adjusted.

[0063] To process the captured infrared images, the infrared camera is connected to an image processing computer either via a network or a mobile storage device. The image processing computer includes units for setting and adjusting the thermal resistor geometry calibration, determining the thermal resistor geometry calibration point positions during experiments, and performing post-processing on the infrared images, as well as image stitching, etc., to precisely process the captured infrared images and obtain accurate thermal discharge temperature field data.

[0064] Example 2:

[0065] This embodiment is an improvement upon the above embodiment, refining the resistance connection method of the geometric calibration points of the thermal resistors. In this embodiment, the thermal resistors at each of the geometric calibration points of the thermal resistors are connected in series ( Figure 4 ) or parallel ( Figure 5 ) or grouped series and parallel ( Figure 6 Connect to an adjustable power supply.

[0066] This embodiment illustrates three connection methods for resistance temperature detectors (RTDs). Series connection is the simplest and lowest cost, but if one RTD fails, the entire resistance system will be paralyzed.

[0067] Parallel connection can solve the problem of one RTD failing and affecting the entire system, but the problem of not being able to control each RTD separately is also present when all RTDs are connected in parallel, just like when all RTDs are connected in series.

[0068] The thermal resistors are connected in series and parallel in sections. For example, two thermal resistors in a row are connected in series, or four thermal resistors in a frame are connected in series. Figure 6 As shown, each group of series-connected thermal resistors is connected to an adjustable power supply, and a voltage or current regulator is set on each branch. This allows for individual adjustment of each group of thermal resistors, enabling control over the color of the thermal resistors in the infrared image according to experimental needs, resulting in clearer and more accurate temperature field calibration. The voltage or current regulator can be an adjustable resistor or a semiconductor device such as a thyristor.

[0069] Example 3:

[0070] This embodiment is an improvement upon the above embodiment, detailing the installation method of the thermal resistance geometric calibration point. In this embodiment, the thermal resistance geometric calibration point is supported by a bracket 301 fixed to the bottom of the temperature drainage test tank. Figure 2 As shown.

[0071] The support is actually a straight rod that is fixed upright at the bottom of the water tank.

[0072] Example 4:

[0073] This embodiment is an improvement upon the above embodiment, refining the installation method of the thermal resistance geometric calibration point. In this embodiment, the thermal resistance geometric calibration point is supported by ropes or steel wires 301 fixed to both sides of the water tank. Figure 1 As shown.

[0074] There are two ways to pull a rope or wire, one is as follows: Figure 1 You can run wires horizontally along the sides of the water tank, or you can run wires along the length of the water tank. Horizontal wires facilitate parallel connection of RTDs, while vertical wires facilitate series connection of RTDs.

[0075] Example 5:

[0076] This embodiment is a method for geometric calibration of laboratory temperature and water volume infrared remote sensing measurement using the device described in the above embodiment. The steps of the method are as follows:

[0077] Step 1, set the geometric calibration points of the thermal resistor: calculate the size of the infrared camera's shooting frame and the overlapping area by shooting height, design the placement of the geometric calibration points of the thermal resistor according to the shooting frame, so that there are marker points near the four corners of each frame, and set up and record the coordinate position of the thermal resistor center point of the geometric calibration points of the thermal resistor.

[0078] The key to successful photography lies in the content captured within the frame. Therefore, the position of the infrared camera and the distance between it and the target are crucial, while the placement of the thermistor calibration points is closely related to the frame size. At least four points should be captured within the frame, ideally near the four corners. Since the positions of these four points within the tank are fixed, and the thermistor must be only one or two pixels in size within the frame, the captured image must accurately determine all positional values. Therefore, the camera position and the location of the thermistor calibration points must be carefully planned to achieve highly precise measurements.

[0079] Step 2, Adjust the RTD geometric calibration point: Adjust the RTD current through the RTD geometric calibration point so that the RTD geometric calibration point can be clearly displayed in the field of view of the infrared camera.

[0080] Because infrared photography is used, the images show the color changes caused by the temperature of objects. To ensure the geometric calibration points of the thermal resistor are clearly visible in the photographs, there needs to be a certain contrast between the calibration and the background. Normally, the water temperature in the photograph is ambient temperature, appearing as a pale blue. Warm water typically appears as a pale red, and the temperature of the mixed water gradually changes from pale blue to green, then to yellow, and finally to red. Therefore, by adjusting the voltage and current through the calibration thermal resistor, the thermal resistor appears as a slightly deeper blue than the ambient temperature water in the infrared photograph. This blue distinguishes it from the pale blue of ambient temperature water while also creating a clear contrast with the color of the warm water and its partially mixed components, resulting in a striking effect in the infrared photograph.

[0081] Step 3, conduct model tests: Flow water through a warm drainage experimental tank to simulate the diffusion pattern of warm drainage under environmental flow dilution, and use an infrared camera to capture the temperature field in the warm drainage experimental tank frame by frame along the length of the tank.

[0082] At the start of the experiment, the water flow in the tank was first turned on to simulate the effect of ambient water flow, such as... Figure 1 , 2 The direction of water flow indicated by the hollow arrow in point 3. Then, activate the warm water discharge to create a warm water diffusion pattern. Afterwards, perform infrared imaging.

[0083] Complete water tank stitching photography can be achieved by scanning sequentially, such as from top to bottom, from left to right, or other orders. Precise stitching of the images is then performed using the precise positions of the various thermal resistor geometric calibration points.

[0084] Step 4: Perform geometric calibration of the infrared image.

[0085] This method achieves high-precision image correction through multi-stage processing, specifically including the following sub-steps:

[0086] Sub-step 1, Preparation of geometric coordinate data for water surface: Determine the geometric coordinate array of each image within the shooting area based on the coordinates of the water surface recording points and the actual shooting range;

[0087] Sub-step 2, Image preprocessing: Distortion correction and orientation normalization are performed on the camera image using calibrated camera parameters. Figure 7 To perform nonlinear distortion correction based on the single-line diagram of the photograph, such as... Figure 7 As shown in the photo, the side walls of the water tank on the left and right sides ( Figure 7 The double lines on both sides of the center (the circular lens of the camera) are distorted, forming an arc shape. Distortion correction and preservation of the complete field of view are addressed in [the following section]. Figure 8 ( Figure 8 (Based on a single-line diagram of the photograph), the two side walls of the water tank become straight lines, but the quadrilateral in the field of view is deformed into an inwardly concave quadrilateral.

[0088] 2.1 Distortion Correction:

[0089] The original image was calibrated using camera parameters. Figure 7 Nonlinear distortion correction is performed while preserving the complete field of view. Figure 8 ).

[0090] Mathematical principles:

[0091] x_corrected=x*(1+k1*r 2 +k2*r 4 +k3*r 6 )+2*p1*x*y+p2*(r 2 +2x 2 )

[0092] y_corrected=y*(1+k1*r 2 +k2*r 4 +k3*r 6 )+p1*(r 2 +2y 2 )+2*p2*x*y

[0093] where k1 - k3 are radial distortion coefficients, p1 - p2 are tangential distortion coefficients, and r 2 = x 2 + y 2 .

[0094] 2.2 Direction normalization:

[0095] Rotate the image according to the direction (backward / forward) of the landmark coordinate distribution:

[0096] When the y coordinate decreases (backward), perform a 180° rotation.

[0097] Sub - step 3, Geometric landmark recognition: Combine automatic detection and manual detection. Automatically detect the geometric center of the circular object in the image as the image coordinates of the landmark through automatic recognition, and use manual point selection to complete the missing points for some points that cannot be automatically recognized;

[0098] Automatic detection:

[0099] Adopt an improved circular detection algorithm:

[0100] 3.1 Divide the matrix into a binary matrix according to a specific value, and by adjusting the segmentation threshold, eliminate all water temperature change intervals, so that the image only shows the graphic objects whose temperature values are in the same interval as the thermistor temperature of the landmark;

[0101] 3.2 Delete all objects containing fewer than a specific number of pixels, that is, noise points. The set value of the specific number of pixels is determined through debugging during the initial processing;

[0102] 3.3 Fill the hole pixels inside the object by object boundary tracking and regional feature analysis, using morphological closing operation;

[0103] 3.4 Calculate the area of the object, and screen for circularity based on the ratio of the area A to the square of the perimeter P (circularity index M = 4πA / P 2 ) for circular screening and recognition, and adjust the M value so that all landmarks can be recognized;

[0104] 3.5 Set a low area threshold Amin and a high threshold Amax, and accurately screen out the landmark pixel cluster objects through double filtering of the object area A (Amin < A < Amax);

[0105] Manual intervention mechanism:

[0106] When the automatic detection of some landmarks fails:

[0107] 3.6 Display the candidate points on the RGB image;

[0108] 3.7 Manually complete the landmarks;

[0109] 3.8 Perform image center calculation on all calibration points and automatically sort and optimize the center coordinates so that the points are arranged according to a specific starting point and rotation direction;

[0110] Sub-step 4, Image Geometric Correction: This is divided into two stages. In the first stage, based on the ratio of the horizontal spacing dx0 to the vertical spacing dy0 represented by the geometric coordinates of the four marker points within the image frame, and the horizontal pixel coordinate range [x1, x2] and vertical pixel coordinate range [y1, y2] represented by the image coordinates of the four marker points, new marker point image coordinates are determined. This ensures that the horizontal and vertical spacing between marker points is not less than the distance between x1 and x2 and the distance between y1 and y2, respectively, and the ratio of the horizontal and vertical distances between marker points is close to dx0 / dy0. Based on the new marker point image coordinates, the geometrically corrected image is bilinearly interpolated and projected onto the new image. The purpose of this step is mainly to correct geometric distortions without significantly changing the image resolution. The second stage involves reprojecting the new image corrected in the first stage according to the geometric coordinates of the four marker points, transforming the image into the result of actual geometric coordinates, thus preparing for the stitching of multiple images.

[0111] Sub-step 5, Image Output: Generate an image file from the image with actual geometric coordinates to provide a basis for subsequent temperature correction between different amplitudes; at the same time, for the convenience of subsequent image stitching, based on the geometric coordinates of the 4 marker points, crop out the image within the quadrilateral area enclosed by the marker points, and output the cropped image as the image stitching output data after geometric correction.

[0112] 5.1 Full-frame output:

[0113] Generate an uncropped image file containing combined geometric coordinate information;

[0114] 5.2 Intelligent cropping:

[0115] Automatically search for the boundary range of the effective rectangular area represented by the calibration point, perform image cropping accordingly, and output cropped image file;

[0116] Sub-step 6, image stitching: After processing multiple images sequentially in the image processing software, the stitched image result is directly output by cropping and stitching the images.

[0117] Finally, it should be noted that the above is only used to illustrate the technical solution of the present invention and not to limit it. Although the present invention has been described in detail with reference to the preferred arrangement, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solution of the present invention (such as the form of the water tank, the simulated water flow movement mode, the processing software used, the order of steps, etc.) without departing from the spirit and scope of the technical solution of the present invention.

Claims

1. A method for laboratory warm drain infrared remote sensing measurement geometric calibration, the method using apparatus comprising: The warm drainage experimental water tank is provided with an infrared camera above the water tank, which can move and shoot along the length direction of the water tank, a plurality of thermal resistance geometric calibration points are arranged at a certain height from the water surface in the water tank and connected with an adjustable power supply, and the arrangement positions of the thermal resistance geometric calibration points are as follows: at least one column of thermal resistance geometric calibration points is distributed on the left and right sides along the width direction of the water tank. The spacing of the thermal resistance geometric calibration points along the length direction of the water tank needs to ensure that at least two rows of thermal resistance geometric calibration points can be accommodated in the up-down direction of the infrared camera frame. The geometric size of the thermal resistance geometric calibration points is required to be at least 1 pixel in the frame of the infrared camera, the thermal resistance of each thermal resistance geometric calibration point is connected with the adjustable power supply in series or parallel or in group series and parallel, and the thermal resistance geometric calibration points are supported by a support fixed on the bottom of the warm drainage experimental water tank or a rope or steel wire fixed on the two sides of the water tank. The method comprises the following steps: Step 1, arranging thermal resistance geometric calibration points: the shooting frame size and the overlapping area of the infrared camera are calculated through the shooting height, the arrangement positions of the thermal resistance geometric calibration points are designed according to the shooting frame, so that there are calibration points near the four corner points in each frame, and the coordinates of the thermal resistance center points of the thermal resistance geometric calibration points are arranged and recorded; Step 2, debugging the thermal resistance geometric calibration points: the current of the thermal resistance of the thermal resistance geometric calibration points is debugged, so that the thermal resistance geometric calibration points can be clearly displayed in the field of view of the infrared camera; Step 3, carrying out model test: the diffusion form of the warm drainage under the environmental flow dilution of the water flowing in the warm drainage experimental water tank is simulated, and the infrared camera shoots the temperature field in the warm drainage experimental water tank along the length direction of the water tank; Step 4, carrying out infrared image geometric calibration: comprising the following sub-steps: Sub-step 1, preparing water surface geometric coordinate data: the geometric coordinate array of each image in the shooting area is determined according to the water surface recording point coordinates and the actual shooting range; Sub-step 2, image preprocessing: the camera is corrected and standardized in direction, the original image is corrected in non-linear distortion by using the calibration camera parameters, and the complete field of view is kept; 2.1 distortion correction: the original image is corrected in non-linear distortion by using the calibration camera parameters; 2.2 direction standardization: the image is rotated according to the direction of the calibration point coordinates; Sub-step 3, geometric calibration point identification: automatic detection and manual detection are combined, the geometric center of the circular object in the image is located as the image coordinates of the calibration point by automatic identification, and manual point selection is used to supplement the points that cannot be automatically identified; Automatic detection: improved circular detection algorithm is adopted: 3.1 the matrix is divided into a binary matrix according to a specific value, the segmentation threshold is adjusted, all water temperature change intervals are eliminated, and the image only displays the graphical objects whose temperature value and the temperature of the calibration point thermal resistance are in the same interval; 3.2 all objects containing less than a certain number of pixels, that is, noise points, are deleted, and the specific pixel number is determined by debugging in the initial processing; 3.3 the object boundary tracking and region feature analysis are adopted, and the morphological closing operation is used to fill the hollow pixels in the object; ​ ​ ​ 3.4 Calculate the area of the object, and screen the circularity index M (M=4πA / P²) based on the ratio of the area A and the square of the perimeter P to identify the circle, and adjust the value of M to make all the calibration points be identified; 3.5 Set the low threshold Amin and the high threshold Amax, and double filter the object area A (Amin < A < Amax) to accurately screen out the calibration point pixel cluster object; Manual intervention mechanism: When part of the calibration points fail to be automatically detected: 3.6 Display the candidate points on the RGB image; 3.7 Manually fill in the calibration points; 3.8 Calculate the center of the circle for all calibration points, and automatically sort and optimize the coordinates of the center to arrange the points in a specific starting point and rotation direction; Sub-step 4, image geometric correction: divided into two stages, the first stage determines the new image coordinates of the four markers according to the ratio of the horizontal distance dx0 and the vertical distance dy0 represented by the geometric coordinates of the four markers, and the horizontal pixel coordinate range [x1, x2] and the vertical pixel coordinate range [y1, y2] represented by the image coordinates of the four markers, so that the horizontal and vertical distances between the markers are not less than the distance between x1 and x2 and the distance between y1 and y2, and the ratio of the horizontal distance and the vertical distance between the markers is close to dx0 / dy0. According to the new image coordinates of the markers, the geometrically corrected image is bilinearly interpolated and projected onto a new image. The purpose of this time is to correct geometric distortion without changing the image resolution; The second stage is to project the new image corrected in the first stage according to the geometric coordinates of the four markers again, so that the image is converted into the actual geometric coordinates, which prepares for the splicing of multiple images; Sub-step 5, image output: generate an image file with actual geometric coordinates, which provides a basis for subsequent temperature correction between different images; At the same time, in order to facilitate subsequent image splicing, according to the geometric coordinates of the four markers, the image within the quadrilateral range surrounded by the markers is cropped, and the cropped image is output as the splicing output data after geometric correction; 5.1 Full image output: Generate an uncropped image file containing geometric coordinate information; 5.2 Intelligent cropping: Automatically search for the boundary range of the effective rectangular area represented by the calibration points, and crop the image according to the boundary range to output a cropped image file; Sub-step 6, image splicing: for multiple images, sequentially process them in image processing software and directly splice the output to obtain a splicing image result.

Images