Air cushion furnace strip floating height detection device, detection method and medium

Abstract

The present application relates to a kind of air cushion furnace strip floating height detection device, detection method and medium, detection device is relative to the side wall window of air cushion furnace setting, detection device includes: the equipment support rod of being set in one side of side wall window;Adjusting base is slidably arranged on the equipment support rod;Camera is rotationally arranged on a adjusting base, and different shooting angles of side wall window are formed by corresponding adjusting base respectively;Auxiliary light source is set in one side of side wall window, and the brightness of strip edge is enhanced by projecting light source to side wall window;With each described camera communication connection, the image of the different shooting angle of camera is collected to extract strip edge data, and then the image processing server of strip height is obtained according to strip edge data.The present application realizes the high-precision, real-time detection feedback of the floating height of strip in air cushion furnace, to enhance the quality control ability of strip in heat treatment process, improve the automation control level of production line.

Description

Technical Field

[0001] This invention relates to the field of air cushion furnace control technology, and in particular to a device, method and medium for detecting the floating height of strip in an air cushion furnace. Background Technology

[0002] High-precision copper alloy strip and high-precision aluminum alloy strip are ideal lightweight raw materials for modern industry. Air cushion continuous heat treatment production line has the advantages of high heating and cooling efficiency, high product surface quality and high temperature control accuracy. It is an indispensable key equipment in the production process of high-end strip. Among them, the air cushion furnace is the main equipment in the production line.

[0003] When strip steel undergoes heat treatment in an air cushion furnace, to ensure the surface quality of the strip, rollers cannot be used to support the material inside the furnace. Relying solely on tension rollers at the front and rear of the furnace would cause the strip to break due to excessive tension, while insufficient tension would cause the strip to sag and scratch the surface. To avoid this, air jet devices are evenly arranged on both the upper and lower sides of the air cushion furnace. The continuous air blown by these devices acts like an air cushion to support the strip, keeping it in a floating state throughout the heat treatment process. Hence, it is called an air cushion furnace.

[0004] Currently, the floating height of the strip mainly relies on manual observation through the viewing window on the side wall of the air cushion furnace. Real-time detection and feedback are not possible, and the detection accuracy is not guaranteed. When equipment and process control problems cause unstable floating, the surface of the strip is easily damaged. If alarm feedback cannot be obtained in time and timely response cannot be carried out, more strip waste will be generated, increasing cost waste, causing production delays, and increasing the probability of equipment failure. Summary of the Invention

[0005] (a) Technical problems to be solved

[0006] In view of the above-mentioned shortcomings and deficiencies of the prior art, the present invention provides a device, method and medium for detecting the floating height of air cushion furnace strip, which solves the technical problem that the floating height of the strip needs to be observed manually and the detection accuracy cannot be guaranteed.

[0007] (II) Technical Solution

[0008] To achieve the above objectives, the main technical solutions adopted by the present invention include:

[0009] In a first aspect, embodiments of the present invention provide a binocular vision-based strip floating height detection device for an air cushion furnace, the detection device being disposed relative to a side wall viewing window of the air cushion furnace, the detection device comprising:

[0010] Equipment support rod, which is installed on one side of the side wall viewing window;

[0011] Two adjustable bases, each of which is slidably mounted on the equipment support rod;

[0012] Two cameras, each of which is rotatably mounted on one of the adjustment bases, and each camera forms different shooting angles toward the side wall window by relying on its corresponding adjustment base;

[0013] An auxiliary light source, disposed on one side of a sidewall window, enhances the brightness of the strip edge by projecting light into the sidewall window; and...

[0014] An image processing server is communicatively connected to each of the cameras. The server extracts strip edge data from images captured by the cameras at different shooting angles, and then determines the strip height based on the strip edge data.

[0015] Optionally, the camera lens is connected to the side wall window via a light shield.

[0016] Optionally, the auxiliary light source is a line laser emitter, which is positioned between the two cameras and emits a laser line width ≤ 5mm.

[0017] Optionally, the camera is an industrial camera, and the specifications of the two industrial cameras are identical.

[0018] Secondly, embodiments of the present invention provide a method for detecting the floating height of air cushion furnace strip based on binocular vision, applied to the detection device described above, comprising:

[0019] After the auxiliary light source projects light onto the side wall window to enhance the brightness of the strip edge, the image processing server receives the images captured by the two cameras and performs the following processing:

[0020] The first type of image acquired by the first camera and the second type of image acquired by the second camera are analyzed and processed respectively to obtain the strip edge data of the first type of image and the strip edge data of the second type of image;

[0021] Based on the strip edge pixel height in the strip edge data of the first type of image and the strip edge pixel height in the strip edge data of the second type of image, the strip imaging height offset angle of the first type of image and the strip imaging height offset angle of the second type of image are obtained.

[0022] Based on the strip imaging height offset angle of the first type of image, the strip imaging height offset angle of the second type of image, and the setting parameters of the air cushion furnace and the camera, the strip height is obtained by finding the intersection of the depth extension lines of the two types of offset angles.

[0023] Wherein, the strip edge pixel height is the offset angle of the strip edge pixel height relative to the horizontal line centered on the camera lens.

[0024] Optionally, the first type of image acquired by the first camera and the second type of image acquired by the second camera are analyzed and processed respectively to obtain the strip edge data of the first type of image and the strip edge data of the second type of image, including:

[0025] The first type of image acquired by the first camera and the second type of image acquired by the second camera are respectively processed by median filtering formula;

[0026] By performing binarization processing on the first and second type images after median filtering, the pixel values ​​at each coordinate position that are greater than or equal to a preset threshold are set to 255, and the pixel values ​​that are less than the preset threshold are set to 0, the strip edge data of the first type image and the strip edge data of the second type image are obtained.

[0027] The median filtering formula is as follows:

[0028] g(u,v)=med{f(u+i,v+j)},(i,j)∈S (1)

[0029] In equation (1), u and v are the horizontal and vertical coordinates on the image, respectively, S is the template window, and i and j are the horizontal and vertical coordinate offsets relative to u and v within the template window S. The process of equation (1) is as follows: select a template window S, and make the template window S slide along each pixel from left to right and from top to bottom along the horizontal and vertical directions of the image. After each slide, sort the pixel gray values ​​within the template window range, and take the middle value within the range to replace the pixel gray value at the center position of the template window S.

[0030] Optionally, the strip imaging height offset angle of the first type of image and the strip imaging height offset angle of the second type of image are calculated based on the strip edge pixel height in the strip edge data of the first type of image and the strip edge pixel height in the strip edge data of the second type of image, including:

[0031] By iterating through all non-zero pixels on the strip edge data of the first type of image and the strip edge data of the second type of image, multiple first-type ordinates and multiple second-type ordinates are obtained.

[0032] The average of multiple first-type ordinates and multiple second-type ordinates is calculated to obtain the strip edge pixel height of the first-type image and the strip edge pixel height of the first-type image.

[0033] Based on the strip edge pixel height of the first type of image and the strip edge pixel height of the second type of image, the strip imaging height offset angle of the first type of image and the strip imaging height offset angle of the second type of image are calculated using the offset angle formula:

[0034]

[0035] Where β is the vertical field of view of the camera relative to the side wall window, h is the vertical imaging height of the camera relative to the side wall window, and H is the pixel height of the first type of strip edge or the pixel height of the first type of strip edge.

[0036] Optionally, based on the strip imaging height offset angle of the first type of image, the strip imaging height offset angle of the second type of image, and the setting parameters of the air cushion furnace and the camera, the strip height is obtained by calculating the intersection of the depth extension lines of the two types of offset angles, including:

[0037] Using the cross-section of the air cushion furnace as a two-dimensional coordinate system, with the Y-axis coinciding with the horizontal center of the air cushion furnace and the X-axis coinciding with the bottom of the air cushion furnace, the following parameters are obtained: the width of the air cushion furnace body, the vertical distance from the lens center of the first camera to the bottom of the air cushion furnace, the vertical distance from the lens center of the second camera to the bottom of the air cushion furnace, the tilt angle of the first camera relative to the horizontal plane, the second camera relative to the horizontal plane, the vertical field of view of the first camera, and the vertical field of view of the second camera.

[0038] Based on the aforementioned setting parameters, the formulas for the first type of depth extension line of the strip imaging height offset angle of the first type of image and the formulas for the second type of depth extension line of the strip imaging height offset angle of the second type of image are constructed.

[0039] By combining the first type of depth extension line formula and the second type of depth extension line formula, the intersection point, i.e., the strip height, is obtained;

[0040] in,

[0041] The formula for the first type of depth extension line is:

[0042]

[0043] The formula for the second type of depth extension line is:

[0044]

[0045] The combined formulas are:

[0046]

[0047] In equations (3), (4) and (5), W is the width of the air cushion furnace body, T1 is the vertical distance from the center of the first camera lens to the bottom of the air cushion furnace, and T2 is the vertical distance from the center of the second camera lens to the bottom of the air cushion furnace. k1 and k2 are both slopes, k1 = tan(θ1), k2 = tan(180°-θ2); x is the offset of the strip edge relative to the center line inside the furnace, and y is the required strip height.

[0048] Optionally, after obtaining the strip height by calculating the intersection of the depth extension lines of the two types of offset angles based on the strip imaging height offset angles of the first type of image, the strip imaging height offset angles of the second type of image, and the setting parameters of the air cushion furnace and the camera, the method further includes:

[0049] The image processing server sends the detected strip height data to the connected air cushion furnace production line control platform, so that the air cushion furnace production line control platform can adjust the strip floating height, thus realizing closed-loop detection and control.

[0050] Thirdly, embodiments of the present invention provide a computer-readable medium having computer-executable instructions stored thereon, wherein the executable instructions, when executed by a processor, implement the steps of a binocular vision-based method for detecting the floating height of a gas cushion furnace strip as described above.

[0051] (III) Beneficial Effects

[0052] The beneficial effects of this invention are as follows: This invention employs a non-contact method for high-precision and high-stability online detection of strip floating height. Utilizing the imaging characteristics of binocular vision in 3D space, it solves the problem of insufficient visual depth that single-camera imaging cannot address. Simultaneously, the use of an auxiliary light source minimizes interference from ambient light, improving detection results. This invention's device and method can replace traditional manual observation methods and, when linked with the strip floating control system, form a closed-loop control, contributing to improved stability of the equipment system's process control. Attached Figure Description

[0053] Figure 1 This is a schematic diagram of the composition of a device for detecting the floating height of a gas cushion furnace strip, provided in an embodiment of the present invention.

[0054] Figure 2 This is a flowchart illustrating a method for detecting the floating height of a gas cushion furnace strip, as provided in an embodiment of the present invention.

[0055] Figure 3 This is a schematic diagram of the specific process of step S1 of the air cushion furnace strip floating height detection method provided in an embodiment of the present invention;

[0056] Figure 4 This is a binary processing effect diagram of a method for detecting the floating height of air cushion furnace strip provided in an embodiment of the present invention;

[0057] Figure 5 This is a schematic diagram of the specific process of step S2 in the method for detecting the floating height of air cushion furnace strip provided in an embodiment of the present invention;

[0058] Figure 6 This is a schematic diagram of the offset angle of a method for detecting the floating height of a gas cushion furnace strip, provided in an embodiment of the present invention.

[0059] Figure 7 This invention provides a schematic diagram of the object distance depth for a method of detecting the floating height of a gas cushion furnace strip, as shown in this embodiment.

[0060] Figure 8 This is a schematic diagram of the specific process of step S3 in the method for detecting the floating height of air cushion furnace strip provided in an embodiment of the present invention;

[0061] Figure 9 This is a schematic diagram of the actual height parameters of the strip in a method for detecting the floating height of a gas cushion furnace strip, provided in an embodiment of the present invention.

[0062] [Explanation of Labels in the Attached Image]

[0063] 11: Equipment support rod; 12: Adjustable base; 13: First camera; 14: Second camera; 15: Light shield; 16: Auxiliary light source; 17: Vertical illumination range of auxiliary light source;

[0064] 21: Air cushion furnace; 22: Strip material; 23: Side wall window. Detailed Implementation

[0065] To better explain and facilitate understanding of the present invention, the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0066] First of all, it is worth noting that the strip 22 is a metal strip that is undergoing heat treatment while floating inside the air cushion furnace 21. The air cushion furnace 21 has air holes at the top and bottom that continuously blow air, allowing the strip 22 to be suspended in mid-air.

[0067] This invention provides a binocular vision-based strip floating height detection device for an air cushion furnace. The detection device is positioned relative to a side wall viewing window of the air cushion furnace 21. The side wall viewing window is made of high-temperature resistant glass and is installed on the side wall of the air cushion furnace 21, typically used by maintenance personnel to visually observe the production status of the strip 22 inside the furnace. The detection device includes: a support rod 11, which is positioned on one side of the side wall viewing window; two adjusting bases 12, each of which is slidably mounted on the support rod 11 and can slide vertically; and two cameras, each of which is rotatably mounted on an adjusting base 12 and forms different shooting angles of the side wall viewing window based on its corresponding adjusting base 12. The machine captures images of the furnace from different angles through the viewing window, achieving a binocular imaging effect. The camera's height, position, and angle can be adjusted via the adjustable base 12 to achieve a suitable shooting position and angle. An auxiliary light source 16 is located on one side of the side wall viewing window, projecting light onto the side wall viewing window to enhance the brightness of the strip 22's edge. An image processing server is also included, communicating with each camera. Based on images captured from different shooting angles, the server extracts edge data of the strip 22, and then determines the height of the strip 22 based on this edge data. The image processing server is connected to the industrial camera via network cable and fiber optic cable to acquire and process images in real time.

[0068] This invention employs a non-contact method for high-precision and high-stability online detection of the floating height of strip 22. Utilizing the imaging characteristics of binocular vision in 3D space, it solves the problem of insufficient visual depth that single-camera imaging cannot address. Simultaneously, the auxiliary light source 16 minimizes interference from ambient light, further enhancing the detection effect. This invention's device and method can replace traditional manual observation methods and, when linked with the strip floating control system, form a closed-loop control, contributing to improved stability of the equipment system's process control.

[0069] To better understand the above technical solutions, exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present invention are shown in the drawings, it should be understood that the present invention can be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that the present invention can be understood more clearly and thoroughly, and that the scope of the present invention can be fully conveyed to those skilled in the art.

[0070] Secondly, the camera lens is connected to the side wall window via a lens hood 15. The lens hood 15 is located at the end of both camera lenses and is made of soft, opaque silicone material. This keeps the area where the lens connects to the glass in a shielded space, preventing external light sources from affecting the camera image acquisition through reflection from the glass.

[0071] Next, the auxiliary light source 16 is a line laser emitter, positioned between the two cameras, specifically vertically between them, with a laser line width ≤ 5mm. The advantages of using a line laser are: ordinary light sources have a large illumination area and dispersed light, illuminating the edge of the strip 22 while simultaneously illuminating other heating, detection, and other devices and parts within the furnace, increasing interference, making image processing more difficult, and prone to false detections; line lasers have high brightness and a small range, generating a small area of ​​high-brightness reflection when the laser line strikes the edge of the strip 22 perpendicularly, which may be beneficial for background information removal during image processing. (See also...) Figure 1 It can be seen that the vertical illumination range 17 of the auxiliary light source is projected radially from the side wall window.

[0072] Furthermore, both the first camera 13 and the second camera 14 are industrial cameras, and their specifications are identical. The first camera 13 and the second camera 14 adopt the same specifications, with the following main parameters defined: imaging width w, imaging height h, and vertical field of view β.

[0073] Based on the above device description, the significance of using binocular vision in this invention lies in the following: In the application scenario of this invention, the strip 22 is suspended inside the air cushion furnace 21, and the center of the width direction of the strip 22 coincides with the center of the width in the cross-section of the air cushion furnace 21. When using an industrial camera to take pictures through the viewing window at a fixed angle, the straight-line distance between the edge position of the strip 22 of different widths and the camera lens—the object distance—is not different, but the imaging effect may be the same, such as... Figure 7 The three strips 22 with different widths and different floating heights shown have the same edge height position in the camera image. Therefore, due to the lack of depth direction information in the image, a single industrial camera cannot detect the height of the strip 22 floating. However, by using two cameras to capture the target from different angles and supplementing the spatial depth information with an algorithm, accurate detection can be achieved.

[0074] Secondly, embodiments of the present invention provide a method for detecting the floating height of air cushion furnace strip based on binocular vision, applied to the detection device described above, such as... Figure 2 As shown, the method includes:

[0075] The auxiliary light source 16 (line laser emitter) and the first camera 13 and the second camera 14 are turned on. The line laser emitter emits a vertical laser line into the furnace to illuminate the edge of the strip 22. At the same time, the first camera 13 and the second camera 14 enter the acquisition state, acquire images, and send the images to the image processing server for processing.

[0076] S1. Analyze and process the first type of image acquired by the first camera 13 and the second type of image acquired by the second camera 14 respectively to obtain the strip 22 edge data of the first type of image and the strip 22 edge data of the second type of image.

[0077] like Figure 3 As shown, step S1 includes:

[0078] S11. The first type of image acquired by the first camera 13 and the second type of image acquired by the second camera 14 are respectively processed by the median filtering formula to eliminate the noise generated by the reflection of floating particles in the furnace. Specifically, it can eliminate particle noise with high brightness but small range.

[0079] S12. By performing binarization processing on the first and second type images, the pixel values ​​at each coordinate position that are greater than or equal to a preset threshold are set to 255, and the pixel values ​​that are less than the preset threshold are set to 0, the edge data of the strip 22 of the first type image and the edge data of the strip 22 of the second type image are obtained.

[0080] The first and second type images, after median filtering, are binarized to remove background information and retain only the edges of the strip 22, which are brightly reflected by laser illumination. Binarization distinguishes the grayscale value of each pixel at each coordinate position in the image according to a threshold v: pixel values ​​greater than or equal to v are set to 255, and pixel values ​​less than v are set to 0. After setting, the grayscale image becomes a binary image with only black and white colors, which is usually used for foreground extraction and background removal. In this invention, since the edges of the strip 22 are bright and reach the maximum brightness value of 255, the v value is directly set to 254. Thus, after binarization, only the edges of the strip 22 are retained in the black and white image, and the rest is a black background, thereby achieving the purpose of edge extraction. The effect is shown in the attached figure. Figure 4 As shown.

[0081] The median filtering formula is as follows:

[0082] g(u,v)=med{f(u+i,v+j)},(i,j)∈S (1)

[0083] In equation (1), x and y are the horizontal and vertical coordinate positions on the image, S is the template window, and i and j are the horizontal and vertical coordinate offsets relative to u and v within the template window S, respectively. The process of equation (1) is as follows: select a (2n+1) template window S (usually 3×3 or 5×5), and slide the template window S along each pixel from left to right and from top to bottom along the horizontal and vertical directions of the image. After each slide, sort the pixel gray values ​​within the template window range, and take the middle value within the range to replace the pixel gray value at the center position of the template window S.

[0084] S2. Based on the edge pixel height of strip 22 in the edge data of strip 22 in the first type of image and the edge pixel height of strip 22 in the edge data of strip 22 in the second type of image, calculate the imaging height offset angle of strip 22 in the first type of image and the imaging height offset angle of strip 22 in the second type of image; the edge pixel height of strip 22 is the offset angle of the edge pixel height of strip 22 relative to the horizontal line with the camera lens as the center.

[0085] like Figure 5 Step S2 includes:

[0086] S21. Traverse all non-zero pixels on the edge data of strip 22 of the first type of image and the edge data of strip 22 of the second type of image to obtain multiple first type ordinates and multiple second type ordinates.

[0087] S22. Perform an average calculation on multiple first-type ordinates and multiple second-type ordinates respectively to obtain the edge pixel height of strip 22 of the first-type image and the edge pixel height of strip 22 of the first-type image.

[0088] For the binarized images 1 and 2, the average value of the height direction of the strip 22 edge is calculated as the final height position of the strip 22 in the image pixels. Specifically, all non-zero pixels in the binarized image are traversed to obtain their ordinate y, and the average is calculated by summing these coordinates. The final center position H of the strip 22 edge height direction is obtained, as shown in the following mathematical formula:

[0089]

[0090] Using the formula above, the 22-pixel heights H1 and H2 of the strip in Image 1 and Image 2 are obtained respectively.

[0091] S23. Based on the edge pixel height of the strip 22 in the first type of image and the edge pixel height of the strip 22 in the second type of image, the imaging height offset angle of the strip 22 in the first type of image and the imaging height offset angle of the strip 22 in the second type of image are obtained by using the offset angle formula. Figure 6 The offset angle is shown, and its formula is:

[0092]

[0093] Where β is the vertical field of view of the camera, and H is the edge pixel height of the first type of strip 22 or the edge pixel height of the first type of strip 22.

[0094] S3. Based on the imaging height offset angle of strip 22 from the first type of image, the imaging height offset angle of strip 22 from the second type of image, and the setting parameters of the air cushion furnace 21 and the camera, the height of strip 22 is obtained by calculating the intersection of the depth extension lines of the two types of offset angles. Since industrial cameras lack depth information, the actual height of strip 22 will differ for the same offset angle depending on the width of strip 22. Figure 7 As shown, taking the first camera 13 as an example, with different object distances and the same offset angle, the actual height of the strip 22 that can be represented is not a unique solution, but a set of solutions, and this set of solutions lies on the depth extension line of the offset angle θ. Therefore, two cameras are needed to form a binocular vision system to mutually verify each other and achieve depth detection, thereby obtaining the true height of the strip 22.

[0095] like Figure 8 As shown, step S3 includes:

[0096] S31. Using the cross-section of the air cushion furnace 21 as a two-dimensional coordinate system, with the Y-axis coinciding with the horizontal center of the air cushion furnace 21 and the X-axis coinciding with the inner bottom of the air cushion furnace 21, obtain setting parameters including the width of the air cushion furnace 21, the vertical distance from the lens center of the first camera 13 to the inner bottom of the air cushion furnace 21, the vertical distance from the lens center of the second camera 14 to the inner bottom of the air cushion furnace 21, the tilt angle of the first camera 13 relative to the horizontal plane, the tilt angle of the second camera 14 relative to the horizontal plane, the vertical field of view of the first camera 13, and the vertical field of view of the second camera 14.

[0097] Using the cross-section of the air cushion furnace 21 as a two-dimensional coordinate system, the Y-axis coincides with the horizontal center of the air cushion furnace 21, and the X-axis coincides with the inner bottom of the air cushion furnace 21. Other key parameters are shown in the appendix. Figure 9 As shown:

[0098] W: Width of the air cushion furnace body;

[0099] T1: Vertical distance from the center of the lens of the first camera 13 to the bottom of the inner wall of the air cushion furnace 21;

[0100] T2: Vertical distance from the center of the lens of the second camera 14 to the bottom of the inner part of the air cushion furnace 21;

[0101] W, T1, and T2 are fixed parameters that can be measured during equipment installation. Among them, W is only used for formula derivation.

[0102] also, Figure 9 The diagram also shows that α is the tilt angle of the first camera relative to the horizontal plane; α' is the tilt angle of the second camera relative to the horizontal plane; and β is the vertical field of view angle of the first and second cameras relative to the side wall window. (Since these parameters are shown in the diagram, the instruction manual can retain their description without including them in the calculations.)

[0103] S32. Based on the set parameters, construct the first-class depth extension line formula for the imaging height offset angle of the strip 22 in the first-class image and the second-class depth extension line formula for the imaging height offset angle of the strip 22 in the second-class image.

[0104] S33. By combining the formulas for the first type of depth extension line and the second type of depth extension line, the intersection point, i.e., the height of strip 22, is obtained.

[0105] in,

[0106] The formula for the first type of depth extension line of the first camera 13 is:

[0107]

[0108] The formula for the second type of depth extension line of the second camera 14 is:

[0109]

[0110] The combined formulas are:

[0111]

[0112] In equations (3), (4) and (5), W is the actual width of the air cushion furnace 21, T1 is the vertical distance from the center of the lens of the first camera 13 to the bottom of the air cushion furnace 21, T2 is the vertical distance from the center of the lens of the second camera 14 to the bottom of the air cushion furnace 21, k1 and k2 are both slopes, k1 = tan(θ1), k2 = tan(180°-θ2); x is the offset of the edge of the strip 22 relative to the center line inside the furnace, and y is the required height of the strip 22.

[0113] After calculation using the formula above, the final height y of strip 22 is obtained.

[0114] After step S3, the following also includes:

[0115] S34. The image processing server sends the detected height data of the strip 22 to the connected air cushion furnace production line control platform, so that the air cushion furnace production line control platform can adjust the floating height of the strip 22, thereby realizing closed-loop detection and control.

[0116] Furthermore, embodiments of the present invention also provide a computer-readable medium storing computer-executable instructions, which, when executed by a processor, implement the steps of a binocular vision-based method for detecting the floating height of air cushion furnace strip 22 as described above.

[0117] In one specific embodiment, the above-described solution of the present invention describes the deployment of the device and method in an air cushion furnace production line. It is known that the width W of the air cushion furnace body is 2400mm, the vertical height of the furnace chamber is 500mm, the diameter of the side wall viewing window is 300mm, and the center of the viewing window height coincides with the center of the furnace chamber height.

[0118] The two industrial cameras have an imaging width w of 1920, a height h of 1080, a vertical field of view β of 86.2°, and a lens radius of 33mm.

[0119] When the detection device is deployed, the tilt angle α of the first camera 13 relative to the horizontal plane is 30°, the tilt angle α' of the second camera 14 relative to the horizontal plane is 30°, the vertical distance T1 from the center of the lens of the first camera 13 to the bottom of the air cushion furnace 21 is approximately (furnace height 500mm / 2 + window diameter 300mm / 2 - lens diameter 33mm*sin(30°)) 371.3mm, and the vertical distance T2 from the center of the lens of the second camera 14 to the bottom of the air cushion furnace 21 is approximately (furnace height 500mm / 2 - window diameter 300mm / 2 + lens diameter 33mm*sin(30°)) 128.7mm.

[0120] If the current strip 22 width is 1600mm, the horizontal center of strip 22 coincides with the horizontal center of the furnace body, and the current suspension height is 260mm, perform the following visual inspection under these conditions:

[0121] First, the image captured by the first camera 13 is filtered and binarized to obtain a pixel height H of 366 pixels at the edge of the strip 22, and θ1 is calculated to be 14.944° according to Formula 3. The image captured by the second camera 14 is filtered and binarized to obtain a pixel height H of 684 pixels at the edge of the strip 22, and θ2 is calculated to be 17.453° according to Formula 3. Next, based on the above known conditions, the slope k1 is calculated to be 0.2669 and k2 to be -0.314 using Formula 6. Then, the height y of the strip 22 is calculated to be 259.91 mm, approximately 260 mm, which matches the actual height of the strip 22, indicating successful detection.

[0122] In summary, this invention provides a device, method, and medium for detecting the floating height of strip in an air cushion furnace. The detection device consists of two industrial cameras, one line laser emitter, adjustable bases 12, and a support frame. The two adjustable bases 12 are mounted on the support frame and can be moved up and down to fine-tune the height. The two industrial cameras are respectively mounted on the two adjustable bases 12, allowing for vertical rotation and angle adjustment. Each industrial camera is equipped with a light shield 15 at its end to prevent glare from the side wall viewing window. The line laser emitter, fixed to the support frame by a support rod, is located between the two cameras and has a vertical movement function for height adjustment. The side wall viewing window is made of high-temperature resistant glass and is installed on the side wall of the air cushion furnace 21. It is typically used by maintenance personnel to visually observe the production status of the strip 22 inside the furnace; this invention replaces the human eye by installing a detection device. In addition to the aforementioned devices, the detection system also requires an image processing server for image processing. The image processing server is connected to an industrial camera via network cable and fiber optic cable to acquire and process images in real time, and then feed the detection results back to the production line control system to form a closed-loop control for detection. Furthermore, this invention designs a method for detecting the floating height of strip material in an air cushion furnace based on binocular vision, which includes: first, enhancing the edge brightness of the strip material 22 with a line laser light source, then capturing images with a camera, performing image processing on the server system, extracting the edge of the strip material 22 in the image, and finally calculating the height of the strip material 22 using an algorithm.

[0123] Therefore, this invention achieves high-precision, real-time detection and feedback of the floating height of the strip in the air cushion furnace, thereby enhancing the quality control capability of the strip during the heat treatment process and improving the automation control level of the production line.

[0124] Since the systems / devices described in the above embodiments of the present invention are systems / devices used to implement the methods of the above embodiments of the present invention, those skilled in the art can understand the specific structure and modifications of the systems / devices based on the methods described in the above embodiments of the present invention, and therefore will not be repeated here. All systems / devices used in the methods of the above embodiments of the present invention fall within the scope of protection of the present invention.

[0125] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0126] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, as well as combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions.

[0127] It should be noted that any reference numerals placed between parentheses in the claims should not be construed as limiting the claims. The word "comprising" does not exclude the presence of components or steps not listed in the claims. The word "a" or "an" preceding a component does not exclude the presence of a plurality of such components. The invention can be implemented by means of hardware comprising several different components and by means of a suitably programmed computer. In claims that enumerate several means, several of these means may be embodied by the same hardware. The use of the terms first, second, third, etc., is merely for convenience of expression and does not indicate any order. These terms can be understood as part of the component names.

[0128] Furthermore, it should be noted that in the description of this specification, the terms "one embodiment," "some embodiments," "embodiment," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Furthermore, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0129] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the claims should be interpreted to include both the preferred embodiments and all changes and modifications falling within the scope of the invention.

[0130] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, then this invention should also include these modifications and variations.

Claims

1. A device for detecting the floating height of air cushion furnace strip based on binocular vision, characterized in that, The detection device is positioned relative to the side wall viewing window of the air cushion furnace, and the detection device includes: Equipment support rod, which is installed on one side of the side wall viewing window; Two adjustable bases, each of which is slidably mounted on the equipment support rod; Two cameras, each of which is rotatably mounted on one of the adjustment bases, and each camera forms different shooting angles toward the side wall window by relying on its corresponding adjustment base; An auxiliary light source, disposed on one side of a sidewall window, enhances the brightness of the strip edge by projecting light into the sidewall window; and... An image processing server is communicatively connected to each of the cameras. The server extracts strip edge data from images captured by the cameras at different shooting angles, and then determines the strip height based on the strip edge data.

2. The binocular vision-based air cushion furnace strip floating height detection device as described in claim 1, characterized in that, The camera lens is connected to the side wall window via a lens hood.

3. The binocular vision-based air cushion furnace strip floating height detection device as described in claim 1, characterized in that, The auxiliary light source is a line laser emitter, which is positioned between the two cameras and emits a laser line width ≤ 5mm.

4. A binocular vision-based device for detecting the floating height of air cushion furnace strip as described in any one of claims 1-3, characterized in that, The cameras are industrial cameras, and the specifications of the two industrial cameras are identical.

5. A method for detecting the floating height of air cushion furnace strip based on binocular vision, applied to the detection device as described in any one of claims 1-4, characterized in that, include: After the auxiliary light source projects light onto the side wall window to enhance the brightness of the strip edge, the image processing server receives the images captured by the two cameras and performs the following processing: The first type of image acquired by the first camera and the second type of image acquired by the second camera are analyzed and processed respectively to obtain the strip edge data of the first type of image and the strip edge data of the second type of image; Based on the strip edge pixel height in the strip edge data of the first type of image and the strip edge pixel height in the strip edge data of the second type of image, the strip imaging height offset angle of the first type of image and the strip imaging height offset angle of the second type of image are obtained. Based on the strip imaging height offset angle of the first type of image, the strip imaging height offset angle of the second type of image, and the setting parameters of the air cushion furnace and the camera, the strip height is obtained by finding the intersection of the depth extension lines of the two types of offset angles. Wherein, the strip edge pixel height is the offset angle of the strip edge pixel height relative to the horizontal line centered on the camera lens.

6. The method for detecting the floating height of air cushion furnace strip based on binocular vision as described in claim 5, characterized in that, The first type of image acquired by the first camera and the second type of image acquired by the second camera are analyzed and processed respectively to obtain the strip edge data of the first type of image and the strip edge data of the second type of image, including: The first type of image acquired by the first camera and the second type of image acquired by the second camera are respectively processed by median filtering formula; By performing binarization processing on the first and second type images after median filtering, the pixel values ​​at each coordinate position that are greater than or equal to a preset threshold are set to 255, and the pixel values ​​that are less than the preset threshold are set to 0, the strip edge data of the first type image and the strip edge data of the second type image are obtained. The median filtering formula is as follows: g(u,v)=med{f(u+i,v+j)},(i,j)∈S (1) In equation (1), u and v are the horizontal and vertical coordinates of the image, respectively, S is the template window, and i and j are the horizontal and vertical coordinate offsets of the template window S relative to u and v. The process of equation (1) is as follows: select a template window S, and make the template window S slide along each pixel from left to right and from top to bottom along the horizontal and vertical directions of the image. After each slide, sort the pixel gray values ​​within the template window range, and take the middle value within the range to replace the pixel gray value at the center position of the template window S.

7. The method for detecting the floating height of air cushion furnace strip based on binocular vision as described in claim 5, characterized in that, Based on the strip edge pixel height in the strip edge data of the first type of image and the strip edge pixel height in the strip edge data of the second type of image, the strip imaging height offset angle of the first type of image and the strip imaging height offset angle of the second type of image are calculated as follows: By iterating through all non-zero pixels on the strip edge data of the first type of image and the strip edge data of the second type of image, multiple first-type ordinates and multiple second-type ordinates are obtained. The average of multiple first-type ordinates and multiple second-type ordinates is calculated to obtain the strip edge pixel height of the first-type image and the strip edge pixel height of the first-type image. Based on the strip edge pixel height of the first type of image and the strip edge pixel height of the second type of image, the strip imaging height offset angle of the first type of image and the strip imaging height offset angle of the second type of image are calculated using the offset angle formula: Where β is the vertical field of view of the camera relative to the side wall window, h is the vertical imaging height of the camera relative to the side wall window, and H is the pixel height of the first type of strip edge or the pixel height of the first type of strip edge.

8. The method for detecting the floating height of air cushion furnace strip based on binocular vision as described in claim 5, characterized in that, Based on the strip imaging height offset angle of the first type of image, the strip imaging height offset angle of the second type of image, and the setting parameters of the air cushion furnace and the camera, the strip height is obtained by calculating the intersection of the depth extension lines of the two types of offset angles, including: Using the cross-section of the air cushion furnace as a two-dimensional coordinate system, with the Y-axis coinciding with the horizontal center of the air cushion furnace and the X-axis coinciding with the bottom of the air cushion furnace, the following parameters are obtained: the width of the air cushion furnace body, the vertical distance from the lens center of the first camera to the bottom of the air cushion furnace, the vertical distance from the lens center of the second camera to the bottom of the air cushion furnace, the tilt angle of the first camera relative to the horizontal plane, the second camera relative to the horizontal plane, the vertical field of view of the first camera, and the vertical field of view of the second camera. Based on the aforementioned setting parameters, the formulas for the first type of depth extension line of the strip imaging height offset angle of the first type of image and the formulas for the second type of depth extension line of the strip imaging height offset angle of the second type of image are constructed. By combining the first type of depth extension line formula and the second type of depth extension line formula, the intersection point, i.e., the strip height, is obtained; in, The formula for the first type of depth extension line is: The formula for the second type of depth extension line is: The combined formulas are: In equations (3), (4) and (5), W is the width of the air cushion furnace body, T1 is the vertical distance from the center of the first camera lens to the bottom of the air cushion furnace, T2 is the vertical distance from the center of the second camera lens to the bottom of the air cushion furnace, k1 and k2 are both slopes, k1 = tan(θ1), k2 = tan(180°-θ2); x is the offset of the strip edge relative to the center line inside the furnace, and y is the required strip height.

9. The method for detecting the floating height of air cushion furnace strip based on binocular vision as described in claim 5, characterized in that, Based on the strip imaging height offset angles of the first type of image and the second type of image, as well as the setting parameters of the air cushion furnace and the camera, after obtaining the strip height by calculating the intersection of the depth extension lines of the two types of offset angles, the process also includes: The image processing server sends the detected strip height data to the connected air cushion furnace production line control platform, so that the air cushion furnace production line control platform can adjust the strip floating height, thus realizing closed-loop detection and control.

10. A computer-readable medium having computer-executable instructions stored thereon, characterized in that, When the executable instructions are executed by the processor, they implement the steps of the binocular vision-based air cushion furnace strip floating height detection method as described in any one of claims 5-9.

Images