A strip explicit shape on-line vision detection system and method for roughing process

Abstract

The present application provides a kind of rough rolling process strip explicit plate shape online visual detection system and method, the present application system includes strip detection system, grating light grid structured light projection system, image acquisition system, distance adjustment system and image processing system;When strip enters hot detector detection range, transmission signal triggers grating light grid structured light projection system and image acquisition system start working;Image acquisition system is used to transmit the image information collected to image processing system in real time;Distance adjustment system is used to adjust the distance between grating light grid structured light projection system and image acquisition system.Image processing system is used to process the image transmitted by image acquisition system in real time, obtains the three-dimensional profile of strip, obtains the two-dimensional curve of strip in XOY coordinate system and the two-dimensional curve in YOZ coordinate system, and then discriminates the degree and type of strip buckling head and camber.

Description

Technical Field

[0001] This invention relates to the field of online real-time detection technology for explicit strip shape in the roughing stage of hot rolling mills in the steel rolling industry. More specifically, it relates to an online visual detection system and method for explicit strip shape in the roughing process. Background Technology

[0002] The nation is increasingly emphasizing industrial development, and the steel industry, as the foundation of industrial development, has always played a crucial role, contributing significantly to warships, bridges, and construction. However, in actual hot strip rolling production, various external factors often cause warping, undercutting, and camber in the rolled strips. This not only interferes with the production quality and efficiency of strip steel but also damages equipment, negatively impacting the development of the entire strip steel production enterprise. Currently, due to the excessively high temperatures during strip steel production, most steel mills in China measure undercutting or warping in roughing strip steel by stopping the machine and waiting for the strip to cool down, then manually measuring the gap between the rod and the steel plate using a ruler (this method is rarely used) or by visual inspection on the production line. This severely restricts production efficiency and is labor-intensive. Therefore, there is a need to develop a new non-contact measurement system for online real-time detection of strip steel.

[0003] Visual measurement technology is a novel measurement technique. Its main research focus is on measuring the geometric dimensions and position of objects. It can be widely applied to active, real-time measurement processes such as online measurement and reverse engineering. The key feature of machine vision systems is increased production flexibility and automation, making them suitable for situations where manual labor is unsuitable.

[0004] Machine vision technology boasts significant advantages such as high inspection speed, high accuracy, and non-contact operation. The non-contact nature of visual inspection eliminates the influence of manual operation and is not time-limited, allowing for continuous operation as long as the equipment is in good working order. Machine vision inspection technology can quickly provide acquired field information to the control center, which will then display the product's outline in real time, laying a solid foundation for automated integrated manufacturing. This inspection technology can reduce inspection costs for enterprises and accomplish various inspection tasks in industrial settings. Since humans are not suited to working in noisy, high-temperature, or other harsh environments, machine vision inspection technology provides reliable protection for such work. When a product changes direction, machine vision inspection technology can quickly detect this change and then inform engineers to make timely adjustments, significantly reducing the occurrence of substandard products.

[0005] In summary, because strip steel is subjected to high temperatures during rough rolling, manual measurement can only be performed after the strip has cooled down. This results in low automation of the production process, and manual inspection is increasingly unable to meet the requirements of modern industry. Adopting machine vision for inspection to improve accuracy, efficiency, and automation has become an urgent need in modern inspection and production. Therefore, machine vision measurement is particularly important for online real-time measurement of strip steel. Due to its advantages such as good reproducibility, high measurement accuracy, high efficiency of non-contact measurement, and low cost, vision measurement technology is a good choice for strip steel contour measurement.

[0006] Machine vision has developed rapidly both domestically and internationally, and is widely used in various fields. Surface inspection based on machine vision includes: PCB printed circuit inspection, STM surface mount inspection, agricultural product quality inspection, road condition inspection, product size inspection, and biopharmaceutical inspection. These inspection systems improve product inspection accuracy and production efficiency.

[0007] However, current machine vision technology faces a bottleneck in online real-time detection and the detection of large objects. Previous research has mostly focused on the detection of small parts or localized parts of objects, which has significant limitations. For online real-time detection of explicit plate shape in hot rolling production lines in the metallurgical industry, domestic steel mills currently rely heavily on manual measurement. Summary of the Invention

[0008] To address the technical problem of the difficulty in achieving real-time online detection of explicit strip shape in roughing rolling processes, this invention provides an online visual inspection system and method for explicit strip shape during roughing rolling. This invention enables real-time detection and analysis of the strip's position and three-dimensional information during the roughing rolling process.

[0009] The technical means employed in this invention are as follows:

[0010] An online visual inspection system for explicit strip shape in roughing rolling processes includes: a strip inspection system, a grating structured light projection system, an image acquisition system, a distance adjustment system, and an image processing system; wherein:

[0011] The strip detection system is connected to the grating grid structured light projection system. When the strip enters the detection range of the thermal detector, the transmission signal triggers the grating grid structured light projection system and the image acquisition system to start working.

[0012] The image acquisition system is connected to the image processing system and is used to transmit the acquired image information to the image processing system in real time.

[0013] The distance adjustment system connects the grating structured light projection system and the image acquisition system, and is used to adjust the distance between the grating structured light projection system and the image acquisition system.

[0014] The image processing system is used to process the images transmitted by the image acquisition system in real time to obtain the three-dimensional contour of the strip, and to obtain the two-dimensional curves of the strip in the XOY coordinate system and the two-dimensional curves in the YOZ coordinate system, thereby determining the degree and type of the strip's buckle and sickle bend.

[0015] Furthermore, the strip inspection system includes a strip, a conveyor roller, and a thermal detector. The strip is placed on the conveyor roller and is conveyed into the inspection range of the thermal detector.

[0016] Furthermore, the grating structured light projection system includes a DLP projector for projecting a grating structured light pattern onto the surface of the conveyor rollers, where the strip is waiting to enter the structured light range.

[0017] Furthermore, the image acquisition system includes a first industrial camera and a second industrial camera. The first and second industrial cameras are symmetrically arranged relative to the center line of the conveyor roller and are equidistant from the DLP projector. The angle between the first and second industrial cameras and the vertical line is 60°, so that the cameras have a certain angle with the side of the strip, which is used to detect the thickness of the strip. The DLP projector projects a structured light pattern in a vertical direction to cover the upper surface of the strip.

[0018] Furthermore, the distance adjustment system includes a slide rail device and an equipment mounting housing; the slide rail device, industrial camera, and DLP projector are all installed inside the equipment mounting housing; the equipment mounting housing is connected to the pedestrian ladder via a U-shaped device and a U-shaped hanging ring; the slide rail device includes a slide rail and a device support rod, the slide rail is fixed on the two inner side walls of the equipment mounting housing, the first industrial camera, the second industrial camera, and the DLP projector are all installed above the crossbeam of the slide rail device via camera mounting plates, and the DLP projector is located at the center of the crossbeam.

[0019] Furthermore, the device housing is also equipped with a cooling device, which includes a water inlet, a sealed wiring hole, and a water outlet on the device housing. It is connected to a water pipe through a water pipe adapter. The device housing is also equipped with a temperature sensor to monitor the device temperature in real time and then send commands to control the water flow rate of the water pipe.

[0020] Furthermore, the online visual inspection system for the strip shape of the roughing process is installed at the inlet and outlet of the roughing mill. The inlet is installed at a certain distance from the roughing mill and is equipped with a walkway ladder. The outlet is installed under the existing frame on site, and both are located directly above the strip conveyor rollers.

[0021] The present invention also provides a method for online visual inspection of the explicit strip shape in the roughing rolling process based on the above-mentioned online visual inspection system for explicit strip shape in the roughing rolling process, comprising:

[0022] The distances between the first industrial camera, the second industrial camera, the DLP projector, and the strip are adjusted using a sliding rail device.

[0023] Adjust the grating structured light pattern projected by the DLP projector so that the edge of the grating structured light pattern coincides with the edge of the ideal strip;

[0024] The improved Zhang Zhengyou calibration algorithm was used to calibrate the online visual inspection system for explicit strip shape in the roughing rolling process, including the intrinsic and extrinsic parameters of the first and second industrial cameras; and the relative position matrices of the first and second industrial cameras and the DLP projector, namely the rotation matrix and translation matrix.

[0025] When the thermal detector detects the strip, it sends a start command, and the first industrial camera, the second industrial camera, and the DLP projector start working.

[0026] The strip movement speed and mill process parameters provided by the steel plant control system are transmitted to the online visual inspection system for strip shape in the roughing process. The first industrial camera and the second industrial camera synchronously capture images in real time, and the shooting frequency is adjusted in real time according to the strip movement speed provided by the steel plant control system to facilitate subsequent image stitching.

[0027] Images captured by the first and second industrial cameras are transmitted to the image processing system for corresponding image preprocessing, including filtering, binarization and thresholding.

[0028] The Canny operator is used to extract 3D feature point clouds from the preprocessed image.

[0029] Match the 3D point cloud of the board at the same moment obtained by the Canny operator from the two cameras. With the y-axis as the dividing line, the first industrial camera uses points with x values ​​from negative to 0, and the second industrial camera uses points with x values ​​from 0 to positive. Match the two results to obtain the complete 3D data of the board at this moment.

[0030] Based on the complete three-dimensional data of the conveyor belt at each moment, since the shooting frequency is based on the conveyor belt speed, the overlapping part of two images is identified, and the last row of data of one image is assigned to the first row of data of the next image, thereby completing the image stitching.

[0031] A surface fitting algorithm is used to fit the stitched image data to obtain a complete three-dimensional contour cloud map of the strip.

[0032] The type and formation process of strip buckle and sickle bend are determined in real time based on the three-dimensional contour cloud map of the strip.

[0033] Compared with the prior art, the present invention has the following advantages:

[0034] 1. The online visual inspection system for explicit strip shape in the roughing process provided by the present invention can detect and analyze the position and three-dimensional information of the strip in real time during the roughing process.

[0035] 2. The online visual inspection system for the explicit shape of strip in the roughing rolling process provided by the present invention uses two cameras, one on the left and one on the right, to achieve online synchronous measurement.

[0036] 3. The online visual inspection system for the visible shape of strip in the roughing rolling process provided by the present invention has a slide rail device that can precisely control the distance between the dual industrial cameras and the DLP projector and the strip.

[0037] 4. The online visual inspection system for strip shape in the roughing rolling process provided by the present invention uses a temperature sensor to monitor the internal and external temperatures of the fixed housing of the equipment in real time, thereby controlling the flow rate of cooling water in real time and ensuring a safe working environment for the inspection system.

[0038] 5. The online visual inspection system for explicit strip shape in the roughing rolling process provided by this invention is a non-contact binocular inspection system for explicit strip shape in hot rolling production lines. It increases the level of intelligence in the production line while ensuring accuracy and efficiency.

[0039] 6. The online visual inspection system for explicit strip shape in the roughing rolling process provided by the present invention has the advantages of real-time performance, high speed and three-dimensional contour visualization, realizing online real-time inspection of explicit strip shape in the hot rolling production line.

[0040] Based on the above reasons, this invention can be widely applied in fields such as online real-time detection of strip shape in the roughing stage of hot rolling mills in the steel rolling industry. Attached Figure Description

[0041] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0042] Figure 1 This is a schematic diagram of the system structure of the present invention.

[0043] Figure 2 This is a schematic diagram of the principle of the binocular detection system provided in an embodiment of the present invention.

[0044] Figure 3 The chessboard calibration plate provided in the embodiments of the present invention.

[0045] Figure 4 This refers to corner detection during the calibration process of the detection system of this invention.

[0046] Figure 5 This is a flowchart of the detection system of the present invention.

[0047] Figure 6 This is a schematic diagram illustrating the image stitching principle provided in an embodiment of the present invention.

[0048] Figure 7 This is a flowchart of the image stitching process provided in an embodiment of the present invention.

[0049] In the diagram: 1. Image processing system; 2. Image acquisition system; 3. Pedestrian ladder; 4. Equipment mounting shell; 5. U-shaped device; 6. Second industrial camera; 7. First industrial camera; 8. DLP projector; 9. Slide rail device; 10. High-transmittance glass; 11. Plate and strip; 12. Conveyor roller; 13. U-shaped lifting ring. Detailed Implementation

[0050] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0051] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit the present invention or its application or use. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0052] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of exemplary embodiments according to the invention. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0053] Unless otherwise specifically stated, the relative arrangement, numerical expressions, and values ​​of the components and steps described in these embodiments do not limit the scope of the invention. It should also be understood that, for ease of description, the dimensions of the various parts shown in the drawings are not drawn to actual scale. Techniques, methods, and devices known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and devices should be considered part of the specification. In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limitations. Therefore, other examples of exemplary embodiments may have different values. It should be noted that similar reference numerals and letters in the following figures denote similar items; therefore, once an item is defined in one figure, it need not be further discussed in subsequent figures.

[0054] In the description of this invention, it should be understood that the orientation or positional relationship indicated by directional terms such as "front, back, up, down, left, right", "horizontal, vertical, horizontal" and "top, bottom" is generally based on the orientation or positional relationship shown in the accompanying drawings, and is only for the convenience of describing this invention and simplifying the description. Unless otherwise stated, these directional terms do not indicate or imply that the device or element referred to must have a specific orientation or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation on the scope of protection of this invention. The directional terms "inner" and "outer" refer to the inner and outer contours relative to the outline of each component itself.

[0055] For ease of description, spatial relative terms such as "above," "over," "on the upper surface of," "above," etc., are used herein to describe the spatial positional relationship of a device or feature as shown in the figures to other devices or features. It should be understood that spatial relative terms are intended to encompass different orientations in use or operation besides the orientation of the device as described in the figures. For example, if the device in the figures is inverted, a device described as "above" or "above" other devices or structures would subsequently be positioned as "below" or "under" other devices or structures. Thus, the exemplary term "above" can include both "above" and "below." The device may also be positioned in other different ways (rotated 90 degrees or in other orientations), and the spatial relative descriptions used herein will be interpreted accordingly.

[0056] Furthermore, it should be noted that the use of terms such as "first" and "second" to define components is merely for the purpose of distinguishing the corresponding components. Unless otherwise stated, the above terms have no special meaning and therefore should not be construed as limiting the scope of protection of this invention.

[0057] like Figure 1As shown, this invention provides an online visual inspection system for the explicit shape of strip in the roughing mill. The system is installed at the entrance and exit of the roughing mill, with the inspection system located directly above the conveyor roller table 12, and the centerline of the inspection device coinciding with the centerline of the conveyor roller table. The system includes: a strip inspection system, a grating structured light projection system, an image acquisition system, a distance adjustment system, and an image processing system; wherein:

[0058] The strip detection system is connected to the grating structured light projection system. When the strip enters the detection range of the thermal detector, the transmission signal triggers the grating structured light projection system to start working.

[0059] The grating structured light projection system is connected to the image acquisition system. When the strip enters the structured light range, the transmission signal triggers the image acquisition system to start working.

[0060] The image acquisition system is connected to the image processing system and is used to transmit the acquired image information to the image processing system in real time.

[0061] The distance adjustment system connects the grating structured light projection system and the image acquisition system, and is used to adjust the distance between the grating structured light projection system and the image acquisition system.

[0062] The image processing system is used to process the images transmitted by the image acquisition system in real time to obtain the three-dimensional contour of the strip, and to obtain the two-dimensional curves of the strip in the XOY coordinate system and the two-dimensional curves in the YOZ coordinate system, thereby determining the degree and type of the strip's buckle and sickle bend.

[0063] In a specific implementation, as a preferred embodiment of the present invention, the strip detection system includes a strip 11, a conveyor roller 12, and a thermal detector. The strip 11 is disposed on the conveyor roller 12 and is conveyed by the conveyor roller 12 into the detection range of the thermal detector.

[0064] In a specific implementation, as a preferred embodiment of the present invention, the grating structured light projection system includes a DLP projector 8, which is used to project a grating structured light pattern onto the surface of the conveyor roller 12 to wait for the strip 11 to enter the range of the structured light.

[0065] In a specific implementation, as a preferred embodiment of the present invention, the image acquisition system includes a first industrial camera 7 and a second industrial camera 6. The first industrial camera 7 and the second industrial camera 6 are symmetrically arranged relative to the center line of the conveyor roller 12 and are respectively equidistant from the DLP projector 8. The angle between the first industrial camera 7 and the second industrial camera 6 and the vertical line is 60°, so that the camera has a certain angle with the side of the strip 11, which is used to detect the thickness of the strip 11. The DLP projector 8 projects a structured light pattern in a vertical direction to cover the upper surface of the strip 11.

[0066] In a specific implementation, as a preferred embodiment of the present invention, the distance adjustment system includes a slide rail device 9 and an equipment fixing housing 4; the slide rail device 9, the industrial camera, and the DLP projector 8 are all disposed inside the equipment fixing housing 4; the equipment fixing housing 4 is connected to the pedestrian ladder 3 through a U-shaped device 5 and a U-shaped hanging ring 13; the slide rail device 9 includes a slide rail and a device support rod, the slide rail is fixed on the two inner side walls of the equipment fixing housing, the first industrial camera 7, the second industrial camera 6, and the DLP projector 8 are all disposed above the crossbeam of the slide rail device 9 through a camera fixing plate, and the DLP projector 8 is located at the center of the crossbeam. The slide rail device 9 is used to adjust the distance between the first industrial camera 7, the second industrial camera 6, and the DLP projector 8 and the conveyor roller 12, so that the detection system and the strip 11 can be kept within an effective detection range and other external environmental interference can be reduced. In this embodiment, the first industrial camera 7 and the second industrial camera 6 are symmetrical with respect to the center line of the detection system and are 0.9m away from the DLP projector 8 respectively. The angle between the first industrial camera 7 and the second industrial camera 6 and the vertical line is 60°, so that the camera has a certain angle with the side of the strip, which can detect the thickness of the strip. The structured light pattern projected by the DLP projector is vertical and covers the upper surface of the strip.

[0067] In a specific implementation, as a preferred embodiment of the present invention, the device fixing housing 4 is also provided with a cooling device. The cooling device includes a water inlet hole, a sealed wiring hole and a water outlet hole opened on the device fixing housing 4, and is connected to a water pipe through a water pipe conversion connector. The device fixing housing 4 is also provided with a temperature sensor to monitor the temperature of the device in real time, and then send a command to control the water flow rate of the water pipe.

[0068] like Figure 2The diagram shows the workflow of the online visual inspection system for explicit strip shape in the roughing rolling process of this invention. Before inspection begins, the system is turned on and then calibrated to determine the intrinsic and extrinsic parameters of the dual industrial cameras and the projection matrix M of the dual cameras. Then, the strip enters the inspection system, which begins image acquisition. Image processing system 1 receives the images transmitted in real-time from image acquisition system 2, performs image preprocessing and edge contour extraction, and then performs three-dimensional feature point extraction. Data fusion is performed on the extracted three-dimensional feature points from the images captured by the dual cameras. The last row of data from the fused image is assigned to the first row of the next image, resulting in a complete image stitching result. This automatically identifies overhangs and cambers in the explicit strip shape.

[0069] The present invention also provides a method for online visual inspection of the explicit strip shape in the roughing rolling process based on the above-mentioned online visual inspection system for explicit strip shape in the roughing rolling process, comprising:

[0070] Adjust the distance between the first industrial camera 7, the second industrial camera 6, and the DLP projector 8 and the conveyor roller 12, and use the cross section of the conveyor roller 12 as the reference plane.

[0071] The system calibration uses an improved version based on Zhang Zhengyou's calibration algorithm to calibrate the first industrial camera 7 and the second industrial camera 6, determining the camera's intrinsic and extrinsic parameters: industrial camera focal length f, CCD tilt factor γ, camera center position (u0, v0), imaging point pixel coordinates (u, v), and world coordinate system (X). w Y w Z w ), scale factor s, camera intrinsic parameters A, rotation matrix R, translation matrix T, projection matrix H.

[0072] H = A[RT]

[0073]

[0074] Due to the presence of the protective glass, an object point offset Δ was introduced in the improvement. M1 is the internal parameter matrix of the industrial camera without protective glass, M2 is the external parameter matrix of the industrial camera without protective glass, and M3 is the glass refraction influence matrix.

[0075]

[0076] The calibration model described above yields the intrinsic and extrinsic parameters of the first industrial camera 7 and the second industrial camera 6, as well as the positional geometric relationship between them.

[0077] The camera calibration toolbox in MATLAB software was further developed. Keeping the camera stationary, the positions of the calibration board were changed to allow the first industrial camera 7 and the second industrial camera 6 to capture images. Each camera captured ten sets of images, and then the intrinsic and extrinsic parameters, distortion coefficients, projection matrix H, and positional parameters between the two industrial cameras were solved.

[0078] Images captured by the first industrial camera 7 and the second industrial camera 6 are transmitted to an image processing system for preprocessing, including filtering, binarization, and thresholding. The preprocessing includes filtering, binarization, and thresholding to obtain an image containing only the strip information. By processing the images captured by the dual industrial cameras, a 3D point cloud is extracted from the images using triangulation and solving for the deformation of the structured light pattern. The point cloud data set extracted by the first industrial camera is P1(t1)......P1(t...). n The point cloud data set extracted by the second industrial camera is P2(t1)......P2(t). n Data fusion mainly involves extracting points from the point cloud data P1(t1) of the first industrial camera at time t1, where the x-values ​​range from negative to 0, and extracting points from the point cloud data P2(t1) of the second industrial camera at time t2, where the x-values ​​range from 0 to positive. These two new datasets are then fused to obtain the complete 3D feature point cloud P(t1) of the strip at time t1, and so on, to obtain P(t1)...P(t...). n );

[0079] For image data stitching, to obtain the complete 3D feature point cloud P1(t) of the board at time t1 obtained by data fusion, the first input is the data set P(t1):

[0080]

[0081] The last row of data in P(t1) is then used as the first row of data in the dataset P(t2) at time t2, resulting in:

[0082]

[0083] By recursively applying this process, the three-dimensional data of the entire strip is finally stitched together to obtain the final three-dimensional data P(t) of the strip.

[0084] By obtaining the two-dimensional curves of the strip in the XOY coordinate system and the YOZ coordinate system, the degree and type of sickle bend are determined based on the difference between the two-dimensional curve in the XOY coordinate system and the original conveyor roller. The two-dimensional curve in the YOZ coordinate system is used to determine the degree of warping and buckling of the buckle head, thereby identifying the degree and type of buckling and sickle bend of the strip.

[0085] The principle of the binocular detection system is as follows: Figure 3 As shown:

[0086] Ω c1 First industrial camera coordinate system O c1 -X c1 Y c1 Z c1 , origin O c1 Located at the optical center of the lens, Z c1 The axis coincides with the camera's optical axis; X c1 Axis, Y c1 The axes are parallel to the horizontal and vertical axes of the imaging plane, respectively.

[0087] Ω c2 First industrial camera coordinate system O c2 -X c2 Y c2 Z c2 , origin O c2 Located at the optical center of the lens, Z c2 The axis coincides with the camera's optical axis; X c2 Axis, Y c2 The axes are parallel to the horizontal and vertical axes of the imaging plane, respectively.

[0088] o-xy: Image coordinate system; uv: Pixel coordinate system, where the x-coordinate u and y-coordinate v of a pixel are the column number and row number in its image array, respectively;

[0089] Ω w World coordinate system O w -X w Y w Z w It is used to represent the spatial position of an object, and is a reference frame in space, the world coordinate system (X-axis) of any point in space. w Y w Z w The Z-axis is parallel to the central axis of the DLP projector 7, and the XOY plane is the reference plane in the measurement model;

[0090] P: Object point, whose projection onto the reference plane is P′, where P is the image point captured by the camera;

[0091] This model allows us to derive the mathematical relationship between pixels and world coordinates:

[0092] f x : The effective focal length of the camera in the x-direction;

[0093] f y : The effective focal length of the camera in the y-direction;

[0094] α: Angle between the structured light plane and the camera's optical axis.

[0095] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A visual online inspection system for the explicit shape of strip during rough rolling, characterized in that, During the rough rolling process of strip, the position and three-dimensional information of the strip are detected and analyzed in real time, including: a strip detection system, a grating structured light projection system, an image acquisition system, a distance adjustment system, and an image processing system; among which: The strip detection system is connected to the grating grid structured light projection system. When the strip enters the detection range of the thermal detector, the transmission signal triggers the grating grid structured light projection system and the image acquisition system to start working. The image acquisition system is connected to the image processing system and is used to transmit the acquired image information to the image processing system in real time. The distance adjustment system connects the grating grid structured light projection system and the image acquisition system, and is used to adjust the distance between the grating grid structured light projection system and the image acquisition system. The distance adjustment system includes a slide rail device and a fixed housing. The slide rail device, industrial camera, and DLP projector are all installed inside the fixed housing. The fixed housing is connected to a pedestrian ladder via a U-shaped device and a U-shaped hanging ring. The slide rail device includes a slide rail and a support rod. The slide rail is fixed to the two inner walls of the fixed housing. The first industrial camera, the second industrial camera, and the DLP projector are all mounted above the crossbeam of the slide rail device via camera mounting plates, with the DLP projector located at the center of the crossbeam. The slide rail device is used to adjust the distance between the first industrial camera, the second industrial camera, the DLP projector, and the conveyor roller, ensuring that the detection system and the strip are within an effective detection range and reducing interference from other external environments. The first and second industrial cameras are symmetrical about the center line of the detection system and are 0.9m away from the DLP projector, respectively. The angle between the first and second industrial cameras and the vertical line is 60°, so that the camera and the side of the strip have a certain angle to detect the thickness of the strip. The DLP projector projects a structured light pattern in a vertical direction, which exactly covers the upper surface of the strip. The device housing is also equipped with a cooling device, which includes a water inlet, a sealed wiring hole and a water outlet on the device housing, and is connected to a water pipe through a water pipe adapter. The device housing is also equipped with a temperature sensor to monitor the temperature of the device in real time and then send commands to control the water flow of the water pipe. The image processing system is used to process the images transmitted by the image acquisition system in real time to obtain the three-dimensional contour of the strip, and to obtain the two-dimensional curves of the strip in the XOY coordinate system and the two-dimensional curves in the YOZ coordinate system, thereby determining the degree and type of the strip's buckle and sickle bend. The workflow of the online visual inspection system for explicit strip shape during the roughing process is as follows: Before starting the inspection, the system is turned on, and then the system is calibrated to determine the intrinsic and extrinsic parameters of the dual industrial cameras and the projection matrix M of the dual cameras. As the strip enters the detection system, the system begins image acquisition. The image processing system receives the images transmitted in real time from the image acquisition system, performs image preprocessing and edge contour extraction, and extracts three-dimensional feature points. It then fuses the three-dimensional feature points from the images captured by the dual cameras. The last row of data from the fused image is assigned to the first row of the next image, thus obtaining a complete image stitching result. The system automatically identifies the raised buckle and sickle-shaped bend of the explicit strip shape. The system calibration uses an improved version based on Zhang Zhengyou's calibration algorithm to calibrate the first and second industrial cameras, determining the cameras' intrinsic and extrinsic parameters: the focal length of the industrial cameras. CCD tilt factor Camera center position Imaging point pixel coordinates World coordinate system , scale factor s Camera internal parameters A, rotation matrix R, translation matrix T, projection matrix H; Due to the presence of the protective glass, an object point offset was introduced in the improvement. , This is the internal parameter matrix of an industrial camera without protective glass. This is the external parameter matrix of an industrial camera without protective glass. The glass refraction effect matrix; The above calibration process yields the intrinsic and extrinsic parameters of the first and second industrial cameras, as well as the positional geometric relationship between them.

2. The online visual inspection system for explicit strip shape in the roughing rolling process according to claim 1, characterized in that, The strip inspection system includes a strip, a conveyor roller, and a thermal detector. The strip is placed on the conveyor roller and is conveyed into the inspection range of the thermal detector.

3. The online visual inspection system for explicit strip shape in the roughing rolling process according to claim 2, characterized in that, The grating structured light projection system includes a DLP projector for projecting a grating structured light pattern onto the surface of the conveyor rollers, where the strip is waiting to enter the range of the structured light.

4. The online visual inspection system for explicit strip shape in the roughing rolling process according to claim 1, characterized in that, The online visual inspection system for the strip shape of the roughing process is installed at the inlet and outlet of the roughing mill. The inlet is installed at a certain distance from the roughing mill and is equipped with a walkway ladder. The outlet is installed under the existing frame on site, and both are located directly above the strip conveyor rollers.

5. A method for online visual inspection of the explicit strip shape in the roughing rolling process based on the online visual inspection system for explicit strip shape in any one of claims 1-4, characterized in that, include: The distances between the first industrial camera, the second industrial camera, the DLP projector, and the strip are adjusted using a sliding rail device. Adjust the grating structured light pattern projected by the DLP projector so that the edge of the grating structured light pattern coincides with the edge of the ideal strip; The improved Zhang Zhengyou calibration algorithm was used to calibrate the online visual inspection system for explicit strip shape in the roughing rolling process, including the intrinsic and extrinsic parameters of the first and second industrial cameras; and the relative position matrices of the first and second industrial cameras and the DLP projector, namely the rotation matrix and translation matrix. When the thermal detector detects the strip, it sends a start command, and the first industrial camera, the second industrial camera, and the DLP projector start working. The strip movement speed and mill process parameters provided by the steel plant control system are transmitted to the online visual inspection system for strip shape in the roughing process. The first industrial camera and the second industrial camera synchronously capture images in real time, and the shooting frequency is adjusted in real time according to the strip movement speed provided by the steel plant control system to facilitate subsequent image stitching. Images captured by the first and second industrial cameras are transmitted to the image processing system for corresponding image preprocessing, including filtering, binarization and thresholding. The Canny operator is used to extract 3D feature point clouds from the preprocessed image. Matching is performed on the 3D point clouds of the board obtained by the dual cameras at the same time using the Canny operator. The first industrial camera is used as the dividing line. x The first industrial camera uses points with values ​​ranging from negative to 0, while the second industrial camera uses points with values ​​ranging from 0 to positive. Matching the two results yields the complete 3D data of the board at that moment. Based on the complete three-dimensional data of the conveyor belt at each moment, since the shooting frequency is based on the conveyor belt speed, the overlapping part of two images is identified, and the last row of data of one image is assigned to the first row of data of the next image, thereby completing the image stitching. A surface fitting algorithm is used to fit the stitched image data to obtain a complete three-dimensional contour cloud map of the strip. The type and formation process of strip buckle and sickle bend are determined in real time based on the three-dimensional contour cloud map of the strip.

Images