A dual-metal steel strip linear calibration detection system based on three-camera grouping measurement

Abstract

The application relates to a bimetal steel strip straightness calibration detection system based on three-camera grouping measurement, relates to the field of metal processing calibration, and comprises a camera measurement module, a first camera, a second camera and a third camera which are arranged at equidistant intervals and are used for respectively collecting edge images of a bimetal steel strip at three different positions; a precision control module which is configured with a camera precision matching strategy and is used for adjusting the detection precision of the first camera, the second camera and the third camera for collecting the edge images; an image optimization module which is configured with a lens distortion calibration strategy and is used for carrying out distortion correction on the images collected by the cameras; and an image processing module which is configured with a deviation adjustment strategy and is used for calculating and analyzing straightness deviation values of the edge images to determine the shearing deviation of the bimetal steel strip and performing feedback adjustment. The application has the effect of improving the shearing size monitoring accuracy of the bimetal steel strip, reducing the time consumption of secondary processing and improving work efficiency.

Description

Technical Field

[0001] This application relates to the field of metal processing calibration technology, and in particular to a bimetallic steel strip linear calibration and testing system based on three-camera group measurement. Background Technology

[0002] Bimetallic steel strip is a composite metal steel strip with excellent physical properties such as high strength, impact resistance, and corrosion resistance. In modern industrial production, bimetallic steel strip is shaped into the required dimensions through shearing.

[0003] In related technologies, during the cutting process of bimetallic steel strips of different sizes and specifications, in order to ensure that the cut size is within the set standard size error range, mechanical devices are used in combination with fiber optic sensors to directly detect and calibrate the bimetallic steel strips. However, when facing detection and calibration work of different processing sizes, fiber optic sensors are difficult to adjust the detection and calibration accuracy, resulting in large detection errors, and secondary processing is required.

[0004] Regarding the aforementioned technologies, the accuracy of fiber optic sensors used in conjunction with mechanical devices for calibration is limited, and they cannot be adapted to and adjusted within a certain accuracy range. This makes it easy for bimetallic steel strips of different sizes and specifications to have errors exceeding the standard requirements during the shearing process, which is not conducive to the accurate processing of products. Furthermore, secondary shearing processing extends the working time and results in low overall work efficiency. Summary of the Invention

[0005] To improve the accuracy of monitoring the shearing dimensions of bimetallic steel strips, reduce the time spent on secondary processing, and increase work efficiency, this application provides a bimetallic steel strip straightness calibration and detection system based on three-camera group measurement.

[0006] In a first aspect, this application provides a bimetallic steel strip linear calibration and detection system based on three-camera group measurement, which adopts the following technical solution:

[0007] A bimetallic steel strip linear calibration and detection system based on three-camera group measurement includes:

[0008] The camera measurement module has a first camera, a second camera, and a third camera arranged at equal intervals along the running path of the bimetallic steel strip after it has been cut, which are used to acquire edge images of the bimetallic steel strip at three different positions respectively.

[0009] The precision control module is connected to the camera measurement module and is configured with a camera precision matching strategy to adjust the detection precision of edge images acquired by the first camera, the second camera and the third camera.

[0010] An image optimization module, connected to the camera measurement module, is configured with a lens distortion calibration strategy, which is used to correct distortion in images acquired by the camera.

[0011] The image processing module is connected with the camera measurement module, and is configured with a deviation adjustment strategy for calculating and analyzing straightness deviation values of the edge image to determine the shearing deviation of the bimetallic steel strip and performing feedback adjustment.

[0012] By adopting the above technical solutions, the modules are cooperated with each other to complete the precision control of camera measurement, lens distortion calibration and deviation adjustment, and other module cooperative work, combined with edge image acquisition and processing, which helps to improve the detection precision and calibration reliability of the straightness of the bimetallic steel strip, adapt to the detection needs of steel strips of different specifications, improve the stability of edge feature recognition, avoid the precision limitation of traditional mechanical calibration, reduce the secondary processing caused by calibration error, and improve the first pass yield and overall efficiency of the steel strip shearing processing.

[0013] Optionally, the deviation adjustment strategy includes:

[0014] According to the edge image, feature recognition is performed to determine position values corresponding to the edge feature positions of the bimetallic steel strip in the first, second and third camera fields of view, and the position values include first, second and third position values;

[0015] The straightness deviation value of the bimetallic steel strip is calculated based on the position values and a preset deviation calculation model, and compared with a preset standard straightness deviation interval to determine a deviation result;

[0016] According to the deviation result, dynamic feedback adjustment is performed on the shearing position of the bimetallic steel strip to keep the straightness deviation value within the preset standard straightness deviation interval.

[0017] By adopting the above technical solutions, the deviation adjustment strategy determines the edge position values through feature recognition, calculates the deviation based on the model and dynamically adjusts, which helps to accurately capture the edge position of the steel strip, realizes quantitative detection and real-time feedback of the straightness deviation, ensures that the deviation value is within the standard interval, avoids the shearing size error caused by excessive deviation, and improves the size consistency of the steel strip shearing.

[0018] Optionally, the deviation calculation model adopts the following formula:

[0019] ;

[0020] ;

[0021] ;

[0022] ;

[0023] wherein, represents the straightness deviation value of the bimetallic steel strip, , , respectively represent the first position value, the second position value and the third position value, K1 is a position compensation coefficient, K2 is a distortion correction coefficient, K3 is a light source influence coefficient, D1 is a lateral deviation value of the actual position of the second group of cameras from the midpoint of the connecting line of the first and third cameras, D2 is a preset reference deviation value, K4 is a preset position influence coefficient, D3 is a difference value between the actual deviation value of the image edge pixel after lens distortion correction and the theoretical non-distortion value, D4 is a preset reference distortion pixel value, K5 is a distortion influence coefficient, D5 is a preset lower limit of the effective uniformity of the light source, D6 is an upper limit of the ideal uniformity of the light source, K3 is a preset light source influence coefficient.

[0024] By adopting the above technical scheme, the deviation calculation model introduces the position compensation coefficient K1, the distortion correction coefficient K2 and the light source influence coefficient K3, and quantifies the compensation of various errors in combination with specific formulas, which helps to comprehensively eliminate the influence of installation position deviation, lens distortion and light source non-uniformity on measurement, improve the calculation precision of the straightness deviation value D, and ensure that the deviation detection result is more in line with the actual situation.

[0025] Optionally, the camera precision matching strategy comprises:

[0026] analyzing according to the straightness deviation value to determine a deviation range, and matching a corresponding lens model in a preset precision database according to the deviation range, wherein the lens model has different shooting field angles;

[0027] determining an optimal shooting lens for the current size specification bimetallic steel strip according to the corresponding lens models of the first camera, the second camera and the third camera, and switching the shooting lens corresponding to the lens model based on the optimal shooting lens.

[0028] By adopting the above technical scheme, the camera precision matching strategy matches the lens model according to the deviation range and switches the optimal lens, which helps to realize the dynamic adaptation of the lens to the size and deviation requirements of the steel strip, avoid the problem of insufficient precision of a single lens in the detection of multiple specifications of steel strips, and improve the adaptability and precision stability of the detection of steel strips of different sizes.

[0029] Optionally, the camera measurement module is further configured with a camera position pre-calibration strategy, comprising:

[0030] ​According to the standard coordinate axis direction formulated according to the movement direction of the bimetallic steel strip, the first camera and the third camera are subjected to coordinate analysis according to the standard coordinate axis direction, and a reference line connected with the coordinates of the first camera and the second camera is generated;

[0031] According to the point-line distance analysis of the reference line and the corresponding coordinates of the second camera located in the middle, the error distance of the second camera and the reference line is determined;

[0032] Based on the error distance, a position compensation coefficient is generated in combination with a preset position compensation model, and the second position value detected by the second camera is corrected according to the position compensation coefficient.

[0033] By adopting the above technical solution, the camera position pre-calibration strategy generates a position compensation coefficient through coordinate analysis and error distance calculation, which helps to eliminate the position error of the camera installation in advance, reduces the interference of the second camera position deviation on measurement, improves the reference consistency of the position value detection, and provides more reliable original data for subsequent deviation calculation.

[0034] Optionally, the lens distortion calibration strategy comprises:

[0035] According to the edge image, sub-pixel level corner point detection is performed to determine the lens distortion parameters of the first camera, the second camera and the third camera;

[0036] Based on the distortion parameters, a polynomial distortion model is constructed, and the real-time collected edge image of the bimetallic steel strip is subjected to reverse coordinate mapping to calculate a pixel level distortion compensation value;

[0037] According to the distortion compensation value, the camera field of view angle and the edge feature position are dynamically corrected, the straight line measurement error after compensation is kept less than a preset upper limit deviation value, and the corrected image is subjected to smoothing processing to keep the edge feature definition.

[0038] By adopting the above technical solution, the lens distortion calibration strategy through sub-pixel corner point detection, polynomial distortion model correction and image smoothing processing helps to accurately eliminate the edge pixel deviation caused by lens distortion, improves the accuracy of edge feature position recognition, ensures that the straight line measurement error is controlled within a preset range, and provides high-quality image data for deviation calculation.

[0039] Optionally, it further comprises a lens field of view enhancement sub-strategy, which adopts the following steps to optimize and adjust the shooting field of view angle of the camera measurement module:

[0040] According to the shearing parameters of the bimetallic steel strip, the corresponding best lens field of view angle in the preset database is matched, and the shooting field of view angles of the first camera, the second camera and the third camera are adjusted;

[0041] The shear parameter change is monitored to determine a trigger view field angle replacement instruction, the optimal view field lens angle is updated according to the changed shear parameter, and the image analysis is carried out according to the updated optimal view field lens angle with a preset verification sub-strategy, so that the image shooting size of the camera is kept within a preset standard size range.

[0042] By adopting the technical scheme, the lens view field enhancement sub-strategy optimizes the view field angle according to the shear parameter and dynamically updates the view field angle, which helps to realize real-time adaptation of the view field angle to the steel strip shear specification, avoid edge missed detection or insufficient precision caused by too large or too small view field, keep the rationality of the image shooting size, and improve the adaptability of multi-specification steel strip detection.

[0043] Optionally, the verification sub-strategy comprises:

[0044] The width pixel range, height pixel range and edge feature definition threshold of the optimal image are determined according to preset standard image size parameters, the standard image size parameters correspond to the shear parameter of the bimetallic steel strip;

[0045] The bimetallic steel strip image under the current view field angle is collected, and the actual width pixel value, actual height pixel value and edge feature definition value of the image are extracted;

[0046] The actual size value is compared with the standard size parameter, the size deviation value and the definition deviation value are calculated, if both are within the preset allowable deviation interval, it is determined that the current view field angle meets the requirements, if not, the camera view field angle is fine-tuned according to the deviation value and the image is re-collected until the image size and definition meet the standard parameter requirements.

[0047] By adopting the technical scheme, the verification sub-strategy compares the size and definition parameters of the actual and standard images and dynamically fine-tunes, which helps to ensure that the image under the current view field angle meets the detection requirements, avoids measurement errors caused by view field deviation, provides a qualified image basis for subsequent edge recognition and deviation calculation, and improves the reliability of the detection result.

[0048] Optionally, the image optimization module is further configured with a light source optimization strategy, comprising:

[0049] The edge image of the bimetallic steel strip is analyzed to determine the edge image shooting area, and the light intensity corresponding to the optimal shooting lens in the preset light database is matched;

[0050] The edge image shooting area is adjusted according to the light intensity and the preset parallel illumination angle, and the light source adjustment gain value is calculated by a preset light source gain analysis model, so that the light source adjustment gain value is kept within a preset reference gain value interval.

[0051] By adopting the technical scheme, the light source optimization strategy matches the light intensity of the edge area and adjusts the gain value, which helps to realize accurate adaptation of the light source to the lens and the edge features of the steel belt, improve the light uniformity and image clarity, avoid edge recognition errors caused by improper lighting, and maintain the stability of the light source adjustment effect.

[0052] Optionally, the light source gain analysis model is calculated by the following formula:

[0053] ;

[0054] wherein, is the light source adjustment gain value, is the image gray scale standard deviation after adjustment, is used to reflect the light uniformity, is the preset image gray scale standard deviation before adjustment, is the edge position detection error, is the preset reference edge error coefficient, is the contrast improvement amount of the image after adjustment, is the preset reference contrast.

[0055] By adopting the technical scheme, the light source gain analysis model quantitatively calculates the gain value G through the formula, combined with the gray scale uniformity, edge error and contrast parameters, which helps to accurately evaluate the light source adjustment effect, realize quantitative control of the gain value, avoid the subjectivity of light source adjustment, improve the accuracy and consistency of light adjustment, and ensure the stability of image quality.

[0056] In summary, the present application includes at least one of the following beneficial technical effects:

[0057] 1. Precision control, lens distortion calibration and deviation adjustment modules work together, combined with edge image acquisition and processing, which helps to improve the detection precision and calibration reliability of the straightness of the bimetallic steel belt, and adapt to the detection needs of steel belts of different specifications, reduce the secondary processing caused by calibration errors, and improve the first pass rate and overall efficiency of steel belt cutting processing;

[0058] 2. The camera position pre-calibration strategy generates a position compensation coefficient through coordinate analysis and error distance calculation, which helps to eliminate the position error of the camera installation in advance, reduces the interference of the second camera position deviation on the measurement, improves the reference consistency of the position value detection, and provides more reliable original data for subsequent deviation calculation;

[0059] 3. The lens distortion calibration strategy uses sub-pixel corner detection, polynomial distortion model correction and image smoothing processing to accurately eliminate the edge pixel deviation caused by lens distortion, improve the accuracy of edge feature position recognition, and ensure that the straight line measurement error is controlled within the preset range, providing high-quality image data for deviation calculation. BRIEF DESCRIPTION OF DRAWINGS

[0060] Figure 1 is a schematic diagram of the connection of system modules in the present application.

[0061] Figure 2 is a method flowchart of steps S100 to S102 in the present application.

[0062] Figure 3 is a method flowchart of steps S200 to S201 in the present application.

[0063] Figure 4 is a method flowchart of steps S300 to S302 in the present application.

[0064] Figure 5 is a method flowchart of steps S400 to S402 in the present application.

[0065] Figure 6 is a method flowchart of steps S403 to S404 in the present application.

[0066] Figure 7 is a method flowchart of steps S4041 to S4043 in the present application.

[0067] Figure 8 is a method flowchart of steps S500 to S501 in the present application. DETAILED DESCRIPTION

[0068] In order to make the purpose, technical solutions and advantages of the present application more clear, the following will combine the drawings of the specification and the examples to further describe the present application in detail. It should be understood that the specific examples described herein are only used to explain the present application, and are not used to limit the present application. Figures 1-8 DETAILED DESCRIPTION

[0069] The embodiments of the present application will be further described in detail below in combination with the drawings of the specification.

[0070] The embodiments of the present application disclose a double metal steel strip straightness calibration detection system based on three-camera grouping measurement. Through mutual cooperation between multiple modules, the precision control of camera measurement, lens distortion calibration and deviation adjustment and other module collaborative work are completed. Combined with edge image acquisition and processing, it is helpful to improve the detection precision and calibration reliability of the straightness of the double metal steel strip, adapt to the detection needs of steel strips of different specifications, improve the stability of edge feature recognition, avoid the precision limitation of traditional mechanical calibration, reduce the secondary processing caused by calibration error, and thus improve the overall efficiency of steel strip shearing processing.

[0071] Referring to Figure 1 The double metal steel strip straightness calibration detection system based on three-camera grouping measurement comprises the following modules:

[0072] The camera measurement module is provided with a first camera, a second camera and a third camera arranged equidistantly along the running path of the bimetallic steel strip after shearing, for collecting edge images of the bimetallic steel strip at three different positions respectively;

[0073] In the embodiments of the present application, the first camera, the second camera and the third camera are industrial area array cameras, such as 12 million pixel CMOS cameras, arranged equidistantly along the horizontal direction of the running direction of the steel strip, the distance between adjacent cameras is 400 mm, the lens is perpendicular to the running direction of the steel strip and faces the edge of the steel strip, to ensure the complete collection of the profile image of the edge of the steel strip. The camera is equipped with a zoom lens with a focal length of 12-35 mm, and is synchronized with the steel strip running speed through a trigger shooting mechanism, for example, when the speed is 1 m / s, the shooting interval is 0.1 s, to avoid image blur. For example, when the bimetallic steel strip runs at a speed of 0.8 m / s, the camera can capture the edge features at different positions in real time, to provide original image data for subsequent straightness calculation.

[0074] The precision control module is connected with the camera measurement module and is configured with a camera precision matching strategy for adjusting the detection precision of the first camera, the second camera and the third camera for collecting edge images.

[0075] The module adjusts through hardware parameters such as exposure time, gain and software algorithm compensation, to ensure that the detection precision of a single camera is stable within ±0.01 mm. The camera precision matching strategy helps to keep the measurement errors of the three cameras consistent, to avoid the straightness calculation error caused by the precision deviation of a single camera, thereby meeting the calibration requirement that the D value straightness deviation is within ±0.03 mm. The specific precision matching strategy is further disclosed and explained in subsequent steps. For example, for bimetallic steel strips of different thicknesses, the module can automatically adjust the camera parameters, to ensure that the subtle deformation of the edge of a thin steel strip can also be accurately captured.

[0076] The image optimization module is connected with the camera measurement module and is configured with a lens distortion calibration strategy for correcting the distortion of the images collected by the camera.

[0077] The module uses multiple interchangeable industrial lenses of different models for image shooting, and corrects the images through a distortion calibration algorithm, such as Zhang's calibration method, to eliminate the influence of lens radial distortion and tangential distortion on edge measurement. At the same time, the module is equipped with a parallel light source, such as an LED strip light source, with adjustable brightness, to irradiate the edge area from the oblique lower side of the steel strip, to enhance the contrast between the edge and the background and reduce the interference of reflected light. The lens distortion calibration strategy helps to improve the authenticity of the edge profile in the image, to provide a reliable image basis for subsequent deviation calculation, and the specific calibration method is further explained in subsequent steps. For example, for the bimetallic steel strip with an arc-shaped edge, the straightness error of the edge in the image can be reduced to within 0.005 mm after distortion correction.

[0078] The image processing module, connected with the camera measurement module, is configured with a deviation adjustment strategy for calculating and analyzing the straightness deviation value of the edge image to determine the shearing deviation of the bimetallic steel strip and for feedback adjustment.

[0079] After receiving the edge images collected by the three cameras, the module extracts the pixel coordinates of the steel strip edge through an edge detection algorithm such as the Canny operator, and converts them into actual physical size pixel-millimeter conversion ratio, for example, according to 1:0.001. Based on the measurement data of the three cameras, the first, second, and third camera measurement width corresponds to the edge position X1, X2, X3, and the straightness deviation value is calculated by the formula X1+X3 / 2-X2 to determine whether the steel strip deviates from the straight line. When the straightness deviation value exceeds ±0.03mm, the module sends an adjustment signal such as a motor drive instruction to the PLC control system to control the calibration mechanical device, and in the embodiment, the steel strip is fine-tuned by the pressure roller. The deviation adjustment strategy helps to correct the straightness deviation of the steel strip after shearing in real time. For example, when the straightness deviation value is detected to be +0.04mm, the steel strip deviates to one side, the system can drive the left pressure roller to slightly pressurize, and the deviation is corrected to within 0.01mm.

[0080] Reference Figure 2 The deviation adjustment strategy includes the following steps:

[0081] Step S100: feature recognition is performed according to the edge image to determine the position values corresponding to the edge feature positions of the bimetallic steel strip in the first, second, and third camera fields of view, the position values including a first position value, a second position value, and a third position value;

[0082] In step S100, feature recognition of the edge image is mainly realized through an image processing algorithm. The system first pre-processes the original image collected by the camera, such as grayscale and denoising, and then extracts the contour features of the bimetallic steel strip edge using an edge detection operator such as the Sobel operator. The edge feature is usually represented as a continuous line with a sudden change in grayscale value in the image, corresponding to the physical boundary of the steel strip edge.

[0083] In specific implementation, the system converts the pixel coordinates of the edge contour into actual physical position values in units of mm: the edge position recognized in the first camera field of view is recorded as the first position value X1, the second camera corresponds to the second position value X2, and the third camera corresponds to the third position value X3. For example, for a bimetallic steel strip with a width of 50mm, the first camera detects the edge position 20.12mm from the left reference line, then X1=20.12, the second camera detects 20.15mm, then X2=20.15, and the third camera detects 20.13mm, then X3=20.13.

[0084] The role of this step is to convert image information into quantifiable physical position data, providing basic parameters for subsequent straightness deviation calculation. By accurately identifying the edge feature position, it ensures that the position value truly reflects the actual running state of the steel strip, such as avoiding edge positioning deviation caused by image blur.

[0085] The feature recognition of the edge image uses a combination algorithm of "preprocessing + adaptive edge detection". The specific execution process is as follows:

[0086] Image preprocessing: First, the color image collected by the camera is processed by grayscale processing. The RGB three-channel pixel value is converted to a single-channel grayscale value by weighted average algorithm, and then the median filter 3x3 filter kernel is used to remove the salt and pepper noise in the image. Dust and water droplets that may exist in the production environment of double-metal steel strip can cause noise in the image. Median filtering can effectively preserve edge information while eliminating isolated noise points.

[0087] Adaptive edge detection: Improved Canny edge detection algorithm is used:

[0088] First, smooth the image by Gaussian filtering to reduce high-frequency noise interference on edge extraction;

[0089] Calculate the image gradient using Sobel operator to get the horizontal and vertical gradient, determine the intensity and direction of the edge;

[0090] Use non-maximum suppression algorithm to thin the edge, only keep the local maximum value pixel in the gradient direction, and compress the wide edge to single-pixel width;

[0091] Based on the gray feature of the steel strip edge, set dynamic double thresholds, such as high threshold T1 = image average gray × 1.2, low threshold T2 = T1 × 0.5, filter out strong edges exceeding T1 and connect weak edges between T1 and T2 and connected with strong edges, and finally get continuous steel strip edge contour.

[0092] Sub-pixel level positioning: Since the steel strip edge precision requirement is high, the camera detection precision is ±0.01mm, based on the pixel level edge, through quadratic curve fitting of the gray distribution of the pixels near the edge, the sub-pixel coordinate positioning precision can reach 0.1 pixel level, and then converted to physical position value, such as pixel size is 0.005mm / pixel, sub-pixel positioning can realize ±0.0005mm theoretical precision.

[0093] For example, for the image of the bimetallic steel strip with slight reflection, the gray value difference of the edge region in the preprocessed gray image is in the range of 50-800-255 gray, and the edge can be accurately extracted by the adaptive Canny algorithm, and after sub-pixel fitting, the error of the edge position values X1, X2, X3 can be controlled within ±0.005 mm, which provides reliable original data for the subsequent straightness deviation calculation.

[0094] The purpose of this step is to convert image information into quantifiable physical position data, and to ensure the anti-interference and precision of edge feature recognition through algorithm optimization. Even in the case of slight scratches and reflections on the surface of the steel strip, the edge position can still be stably extracted, avoiding misalignment caused by feature recognition deviation in subsequent calibration.

[0095] Step S101: Calculate the straightness deviation value of the bimetallic steel strip based on the position value and the preset deviation calculation model, and compare it with the preset standard straightness deviation interval to determine the deviation result;

[0096] In step S101, the deviation calculation model is used to convert the first, second and third position values X1, X2, X3 into a quantitative index representing the straightness of the steel strip, i.e. the straightness deviation value D, so that the running of the steel strip along a straight line can be determined through the correlation of the three positions. The specific content of the model will be further explained in the subsequent steps.

[0097] In specific implementation, the system first calls the deviation calculation model to process the position value and outputs the deviation value D; then compares D with the preset standard straightness deviation interval such as ±0.03 mm: if D is within the interval, it is determined as “deviation qualified”; if D exceeds the interval, such as D=+0.04 mm or D=-0.05 mm, it is determined as “deviation exceeds the standard”. For example, based on the position value in step S100, D=0.01 mm is calculated, which is within the interval of ±0.03 mm, and the deviation result is “qualified”; if another batch of steel strip has D=0.04 mm, the result is “exceeds the standard”.

[0098] The purpose of this step is to determine the straightness state of the steel strip through quantitative comparison, providing a basis for subsequent adjustment. In the qualified state, no intervention is needed, and in the over-standard state, the adjustment mechanism is triggered to ensure that the running deviation of the steel strip is controllable. For example, for bimetallic steel strips used in precision instruments, this step can quickly identify small deviations and avoid product scrap due to cumulative errors.

[0099] Step S102: Dynamically feedback adjust the shearing position of the bimetallic steel strip according to the deviation result, so as to keep the straightness deviation value within the preset standard straightness deviation interval.

[0100] In step S102, the dynamic feedback adjustment is realized through a closed loop of "deviation result-adjustment instruction-executive mechanism": the system converts the deviation result into specific adjustment parameters such as adjustment direction and adjustment amount, and sends them to the control module of the steel strip cutting equipment, such as a PLC system, and the cutting position is corrected by the mechanical executive mechanism, such as a servo motor driven guide roller.

[0101] In specific implementation, if the deviation result is "exceeding the standard" such as D = +0.04 mm, the steel strip deviates to the right side, the system calculates the adjustment amount such as driving the left guide roller to fine-tune 0.02 mm, and sends the instruction to the executive mechanism; after adjustment, the camera re-collects the edge image, and steps S100-S101 are repeated until the deviation value D returns to the interval of ±0.03 mm. For example, a batch of steel strips is caused by mechanical vibration, and D = 0.05 mm, and after the system drives the adjustment mechanism, D = 0.02 mm is obtained again, and the deviation is recovered to be qualified.

[0102] The role of this step is to realize the real-time response of "detection-judgment-adjustment", and to avoid the accumulation of deviation affecting the quality of the steel strip by dynamically correcting the cutting position.

[0103] It needs to be further explained that the deviation calculation model adopts the following formula:

[0104] ;

[0105] ;

[0106] ;

[0107] ;

[0108] Among them, represents the straightness deviation value of the bimetallic steel strip, , , respectively represent the first position value, the second position value and the third position value, is a position compensation coefficient, is a distortion correction coefficient, is a light source influence coefficient, is a transverse deviation value of the actual position of the second group of cameras and the midpoint of the line connecting the first and third cameras, is a preset reference deviation value, is a preset position influence coefficient, is the difference between the actual deviation value of the image edge pixel after lens distortion correction and the theoretical non-distortion value, is a preset reference distortion pixel value, is a distortion influence coefficient, the uniformity of the gray scale of the edge area of the bimetallic steel strip under parallel light source, is a preset lower limit of light source effective uniformity, is a preset upper limit of light source ideal uniformity, is a preset light source influence coefficient.

[0109] Referring to Figure 3 , the camera precision matching strategy comprises:

[0110] Step S200: Analyzing according to the straightness deviation value to determine the deviation range, and matching the corresponding lens model in the preset precision database according to the deviation range, the lens model has different shooting field angles;

[0111] In step S200, the analysis of the deviation range is based on the difference between the straightness deviation value D and the standard interval ±0.03mm: if D is close to the upper limit of the standard interval, such as a small deviation range, it means that the steel strip straightness fluctuation is gentle, and a higher precision lens is needed to capture subtle deviations; if D is far beyond the standard interval, such as a large deviation range, a lens with a larger field of view is needed to avoid the edge of the steel strip from exceeding the shooting range.

[0112] In the preset precision database, the lens models are classified according to the field of view and focal length: wide field of view lenses are suitable for large deviation range, which can cover a larger edge area of the steel strip to avoid the edge from running out of the picture; narrow field of view lenses are suitable for small deviation range with longer focal length, which can capture edge details more clearly and improve positioning accuracy. When executing, the system matches the corresponding lens from the database according to the deviation range, for example, when the steel strip just enters the production line and the deviation range is large, a wide field of view lens is matched; when the deviation is stable in a small range, a narrow field of view lens is automatically matched.

[0113] The purpose of this step is to balance between "complete edge capture" and "high precision measurement" through the adaptation of lens model and deviation range, avoiding missing detection due to insufficient lens field of view, or reducing local measurement accuracy due to excessive field of view.

[0114] Step S201: Determine the optimal shooting lens for the current size specification of the bimetallic steel strip according to the corresponding lens model of the first camera, the second camera and the third camera, and switch the shooting lens corresponding to the lens model based on the optimal shooting lens.

[0115] In step S201, the determination of the optimal shooting lens needs to consider the size specification of the current steel strip and the synergy of the three cameras, including width, thickness, etc. For wide steel strips, the lens field of view needs to be sufficient to cover the full width of the edge; for narrow steel strips, a narrow field of view lens with higher precision can be selected; at the same time, it is necessary to ensure that the lens models of the three cameras are consistent to avoid measurement data deviation caused by lens parameter differences.

[0116] When executed, the system first reads the steel strip size information, such as obtained through the pre-detection module, and then combines the lens model matched in step S200 to screen out the lens that meets the requirements of "covering the edge of the steel strip" and "precision requirement" as the optimal lens. Subsequently, the automatic switching is completed through the lens switching mechanism, such as the electric lens turntable, and after switching, the system will perform short-time calibration, such as shooting the image of the standard plate, to ensure the stability of the lens parameters after switching. For example, when producing wide steel strips, the three cameras are switched to wide field lenses synchronously; when switching to narrow steel strip production, the narrow field high precision lenses are switched synchronously.

[0117] The purpose of this step is to realize the dynamic adaptation of the lens to the steel strip specifications and to ensure the consistency of multi-camera measurement, that is, by unifying the lens model, the measurement deviation of X1, X2 and X3 caused by the difference in the lens of a single camera is avoided, and a unified reference is provided for subsequent straightness calculation.

[0118] Referring to Figure 4 The camera measurement module is also configured with a camera position pre-calibration strategy, including:

[0119] Step S300: According to the motion direction of the bimetallic steel strip, a standard coordinate axis orientation is formulated, and the first camera and the third camera are analyzed according to the standard coordinate axis orientation, and a reference line connecting the coordinates of the first camera and the second camera is generated;

[0120] The motion direction of the bimetallic steel strip refers to the transmission direction of the bimetallic steel strip on the production line, such as horizontal right, which is the basis for formulating the coordinate reference.

[0121] The standard coordinate axis is a three-dimensional coordinate system established with the motion direction of the steel strip as the X axis, the direction perpendicular to the surface of the steel strip as the Z axis, and the horizontal direction perpendicular to the motion direction as the Y axis, which is used to unify the position reference of the camera and the steel strip.

[0122] The reference line is a straight line connecting the Y axis coordinates of the first camera and the third camera in the standard coordinate axis, corresponding to the position in the width direction of the steel strip. In theory, it is a reference line for judging whether the second camera is centered. Because the three cameras are arranged equidistantly, the second camera should be located directly above the midpoint of the reference line.

[0123] When executed specifically, the system first determines the positive direction of the X axis with the motion direction of the steel strip as the reference, and the positive direction of the Z axis as upward perpendicular to the surface of the steel strip to establish the standard coordinate axis; then the coordinates of the first camera and the third camera on the Y axis are obtained through the built-in positioning sensor of the camera, such as the laser positioner, and the reference line is generated by connecting the two coordinates. For example, the Y axis coordinate of the first camera is Y1, the Y axis coordinate of the third camera is Y3, and the reference line is a straight line connecting Y1 and Y3, which is used to judge whether the installation position of the second camera is accurate subsequently.

[0124] The role of this step is to establish a unified coordinate reference system, and to determine the positions of the first and third cameras through the reference line, providing a "reference scale" for subsequent analysis of the position error of the second camera.

[0125] Step S301: Point-line distance analysis is performed on the reference line and the corresponding coordinates of the second camera located in the middle to determine the error distance of the second camera and the reference line.

[0126] Second camera corresponding coordinates: The actual measurement coordinates Y2 of the second camera on the Y-axis in the standard coordinate axis.

[0127] Point-line distance analysis: An analysis method for calculating the vertical distance from the second camera coordinate point Y2 to the first and third camera connecting line of the reference line, used to quantify the installation offset of the second camera.

[0128] Error distance: The result of point-line distance analysis, i.e. the vertical distance between the actual position of the second camera and the reference line, reflecting the position deviation during installation, such as slight misalignment during mechanical installation.

[0129] When executed specifically, the system calls the point-line distance calculation formula based on the geometric algorithm of the standard coordinate system to calculate the Y-axis coordinates Y2 of the second camera and the reference line, and obtains the error distance. For example, if the reference line is a theoretical straight line in the middle, and the actual position of the second camera deviates from the straight line due to installation deviation, the error distance is a numerical value that measures the degree of deviation.

[0130] The role of this step is to quantify the installation error of the second camera. Due to the difficulty of achieving the theoretical center position in actual installation, the error distance can intuitively reflect the size of the deviation, providing a basis for subsequent compensation and correction.

[0131] Step S302: Based on the error distance, a position compensation coefficient is generated based on a pre-set position compensation model, and the second position value detected by the second camera is corrected according to the position compensation coefficient.

[0132] Position compensation model: An algorithm model used to convert error distance into compensation coefficient, the core of which is to establish a corresponding relationship model between "error distance" and "compensation amount". The specific logic is expanded later.

[0133] Position compensation coefficient: A correction parameter calculated based on the error distance, used to adjust the position value detected by the second camera to offset the influence of installation error.

[0134] Second position value correction: The original position value detected by the second camera is combined with the compensation coefficient to obtain the true position value after eliminating the error.

[0135] In specific implementation, the system inputs the error distance into a position compensation model to generate a corresponding compensation coefficient, e.g., the greater the error distance, the greater the absolute value of the compensation coefficient; and then uses the coefficient to correct the second position value X2 detected by the second camera. For example, if the second camera causes the detection value to be smaller due to installation deviation to the left, the compensation coefficient will cause the corrected X2 to be adjusted to the right, close to the real position.

[0136] This step eliminates the influence of installation error on measurement through compensation correction, ensures that the position value of the second camera can truly reflect the edge state of the steel strip, avoids the calculation of straightness deviation value D from being inaccurate due to the deviation of the camera position, and finally improves the overall calibration accuracy.

[0137] Referring to Figure 5 The lens distortion calibration strategy includes:

[0138] Step S400: performing sub-pixel level corner detection on the captured edge image to determine the lens distortion parameters of the first camera, the second camera and the third camera;

[0139] Sub-pixel level corner detection is a feature positioning method with higher precision than pixel level detection. By fitting the gray scale distribution of the pixels around the corner point, the corner point position is positioned to the inside of the pixel rather than the integer coordinates of the pixel, achieving micron-level positioning accuracy. Lens distortion parameters are parameters that describe the degree of deviation of the edge from the ideal straight line during lens imaging, including radial distortion, such as the outward bulging or inward sinking of the edge, and tangential distortion, such as the deviation caused by the tilt of the imaging plane. These parameters are determined by the optical properties of the lens.

[0140] In specific implementation, the system first captures an image of a calibration plate with a standard checkerboard pattern. The checkerboard corner points are feature points with known accurate positions. The actual coordinates of the corner points in the image are extracted through sub-pixel level corner detection, and then compared with the theoretical coordinates of the calibration plate corner points to calculate the deviation between them. The deviation data is a quantitative representation of lens distortion, and the distortion parameters are determined accordingly. For example, a common industrial lens may have slight radial distortion, causing the straight lines in the edge image to appear slightly curved. This step can accurately capture such distortion characteristics.

[0141] The purpose of this step is to provide "raw data" for distortion correction. By detecting and quantifying the distortion parameters, the deviation law of lens imaging is determined, laying a foundation for subsequent image correction.

[0142] Step S401: constructing a polynomial distortion model based on the distortion parameters, performing reverse coordinate mapping on the real-time collected double-metal steel strip edge image, and calculating the pixel-level distortion compensation value;

[0143] The polynomial distortion model is a mathematical model for fitting the lens distortion law with a polynomial function. By substituting the distortion parameters into a polynomial such as a quadratic or cubic polynomial, a corresponding relationship between the "ideal coordinates-distorted coordinates" is established. The reverse coordinate mapping is a reverse calculation process according to the distortion model. Given the coordinates of a pixel in the distorted image, the corresponding coordinates of the pixel in the ideal non-distorted image are derived through the model to realize the reverse correction of the coordinates. The pixel-level distortion compensation value is a value used to correct the position of a single pixel, i.e., the difference between the distorted pixel coordinates and the ideal coordinates, which is used to "pull back" the pixel in the distorted image to the correct position.

[0144] In specific implementation, the system inputs the distortion parameters determined in step S400 into the polynomial distortion model to generate a distortion correction formula. When the steel strip edge image is collected in real time, the coordinates of each pixel in the image are calculated by reverse mapping to obtain the corresponding compensation value. For example, a certain pixel deviates from the ideal position by 0.5 pixels due to radial distortion, and the compensation value is -0.5 pixels, which is used to correct the pixel to the correct position.

[0145] This step is to convert the distortion parameters into an executable correction scheme. Through the mathematical model and coordinate mapping, the compensation amount of each pixel is quantified to provide a calculation basis for the accurate correction of the image.

[0146] Step S402: dynamically correct the camera field of view angle and the edge feature position according to the distortion compensation value, keep the straight line measurement error after compensation less than a preset upper limit deviation value, and perform smoothing processing on the corrected image to keep the edge feature definition.

[0147] The camera field of view angle is the range angle that can be shot by the camera. The distortion will cause a deviation between the actual field of view angle and the theoretical value. The dynamic correction is to adjust the calculation parameters of the field of view angle according to the compensation value to ensure that the imaging range is consistent with the actual situation. The edge feature position is the coordinate of the steel strip edge in the image. The dynamic correction adjusts the coordinate through the compensation value to eliminate the position deviation caused by the distortion. The straight line measurement error is the deviation of the straight line of the steel strip edge in the corrected image from the ideal straight line. The upper limit deviation value is the maximum error allowed by the system, such as ±0.01 mm. The smoothing processing is a noise reduction processing on the corrected image, such as Gaussian blur, to reduce the pixel jump that may be generated in the correction process, while keeping the clear outline of the edge.

[0148] In specific implementation, the system adjusts the edge feature position coordinate and the field of view angle parameter with the compensation value, and then calculates the straight line measurement error after correction. If the error is greater than the upper limit deviation value, the compensation value is re-optimized. If the error meets the standard, the image is smoothed to remove the noise remaining after correction. For example, the straight line measurement error of a certain steel strip edge after correction is 0.008 mm, which is less than the upper limit 0.01 mm. After smoothing processing, the edge outline is continuous and free of sawtooth, which can be directly used for subsequent deviation calculation.

[0149] The role of this step is to control the influence of distortion on measurement within the allowable range through dynamic correction and optimization, while ensuring image quality, ensuring the accuracy of subsequent edge feature recognition and deviation calculation.

[0150] With reference to Figure 6 The lens distortion calibration strategy also includes a lens field of view enhancement sub-strategy, which uses the following steps to optimize and adjust the shooting field of view angle of the camera measurement module:

[0151] Step S403: According to the shearing parameter of the bimetallic steel strip, match the corresponding best lens field of view angle in the preset database, and adjust the shooting field of view angle of the first camera, the second camera and the third camera;

[0152] Specifically, the system first acquires the shearing parameter of the current steel strip, and based on the corresponding relationship between the shearing parameter and the field of view angle in the preset database, matches the best lens field of view angle that can completely cover the edge of the steel strip and retain a reasonable redundancy. Subsequently, through the electric adjustment component of the lens, the shooting field of view angles of the three cameras are synchronously adjusted to the matching value - for example, for a wide steel strip, a larger field of view angle is matched to avoid the edge exceeding the picture; for a narrow steel strip, a relatively smaller field of view angle is matched to reduce invalid background interference.

[0153] The role of this step is to ensure that the camera can always shoot complete edge features through the adaptation of the field of view angle to the steel strip specification, providing effective original images for subsequent distortion calibration and edge recognition.

[0154] Step S404: Monitor the change of the shearing parameter to determine the trigger of the field of view angle replacement instruction, update the best field of view lens angle according to the changed shearing parameter, and perform image analysis according to the updated best field of view lens angle with a preset verification sub-strategy, so that the image shooting size of the camera is kept within a preset standard size range.

[0155] Updating the best field of view lens angle means matching the corresponding field of view angle from the database according to the new shearing parameter. The preset verification sub-strategy is an analysis method for checking whether the adjusted field of view angle is appropriate, which is verified by detecting whether the edge of the steel strip in the image is complete and whether the distance from the edge to the edge of the picture meets the standard, such as retaining at least 5% of the picture width as redundancy. The image shooting size refers to the pixel size of the steel strip area in the image, and the standard size range is a size interval that ensures the accuracy of edge recognition, such as the steel strip area occupying 60%-80% of the image width.

[0156] In specific implementation, if the monitoring detects that the width of the steel strip changes from 50 mm to 70 mm with a change of 20 mm > threshold 10 mm, the system triggers a replacement instruction, and updates the optimal field of view angle to 40°; after adjustment, the image is analyzed by the verification sub-strategy - if the edge of the steel strip is complete and the redundant distance meets the standard, it is confirmed that the adjustment is effective; if the edge part exceeds the picture, the field of view angle is fine-tuned again until the image shooting size is within the standard range.

[0157] The role of this step is to realize the dynamic adaptation of the field of view angle, and through monitoring the parameter change and verifying the adjustment effect, it is ensured that when the specification of the steel strip is switched, the camera can still stably shoot the qualified edge image, providing a reliable image source for subsequent distortion calibration and deviation calculation.

[0158] Referring to Figure 7 The verification sub-strategy includes:

[0159] Step S4041: determining the width pixel range, height pixel range and edge feature definition threshold of the optimal image according to the preset standard image size parameter corresponding to the cutting parameter of the bimetallic steel strip;

[0160] The standard image size parameter refers to the image size reference set to ensure the edge recognition accuracy, which is defined in advance according to the cutting parameters of the bimetallic steel strip, such as width and thickness - for example, a wide steel strip corresponds to a larger width pixel range, and a narrow steel strip corresponds to a relatively small range, so as to ensure that the steel strip area in the image is neither compressed nor redundant. The width pixel range and the height pixel range are specific manifestations of the standard image size parameter, which respectively limit the pixel number interval of the image in the horizontal and vertical directions, such as a width pixel range of 800-1200 pixels and a height pixel range of 500-800 pixels. The edge feature definition threshold is a reference value for measuring whether the edge contour is clear, which is quantified by the edge gradient value, such as a threshold ≥ 80, and the higher the gradient value, the clearer the edge.

[0161] In specific implementation, the system calls the corresponding standard image size parameter according to the cutting parameter of the current steel strip, such as width 60 mm, to determine that the optimal image needs to meet the width pixel 800-1000, the height pixel 500-600, and the edge definition threshold ≥ 85, providing a judgment standard for subsequent image verification.

[0162] The role of this step is to establish the "qualified standard" of image quality, and through the quantification range of size and definition, it is ensured that the subsequent verification has a clear basis.

[0163] Step S4042: acquiring the image of the bimetallic steel strip under the current field of view angle, and extracting the actual width pixel value, the actual height pixel value and the edge feature definition value of the image;

[0164] The actual width pixel value is the total number of pixels in the horizontal direction of the current captured image, such as 900 pixels actually measured, and the actual height pixel value is the total number of pixels in the vertical direction, such as 550 pixels actually measured, both of which reflect the actual size of the image. The edge feature clarity value is a quantitative value obtained by an image algorithm, such as calculating the edge gradient intensity, such as 90 actually calculated, for measuring the clarity of the steel strip edge in the current image.

[0165] In specific implementation, the system controls the camera to take a steel strip image under the current field of view, and automatically extracts the above three actual values, such as image width pixels 920, height pixels 580, and edge clarity value 88, through image processing tools, to provide actual data for subsequent comparison.

[0166] The purpose of this step is to obtain the actual quality parameters of the current image as the basis for comparison with the standard parameters.

[0167] Step S4043: Based on the actual size value and the standard size parameter, the size deviation value and the clarity deviation value are calculated, and if both are within the preset allowable deviation interval, it is determined that the current field of view angle meets the requirements; if not, the camera field of view angle is adjusted according to the deviation value and the image is reacquired until the image size and clarity meet the standard parameter requirements.

[0168] The size deviation value is the difference between the actual width / height pixel value and the standard range, such as actual width 920 pixels, standard range 800-1000 pixels, and deviation value 20 pixels within the qualified range. The clarity deviation value is the difference between the actual edge feature clarity value and the threshold value, such as actual 88 and threshold 85, and deviation value 3. The allowable deviation interval is the preset acceptable deviation range, such as size deviation ≤ ±100 pixels and clarity deviation ≥-5, to ensure that the image quality is within a reasonable fluctuation range.

[0169] In specific implementation, the system compares the actual value with the standard parameter: if both the size deviation and the clarity deviation are within the allowable interval, it is determined that the current field of view angle is qualified; if the actual width pixel 1300 exceeds the upper limit of the standard 100 pixels, the deviation value 200 is calculated, the lens is controlled to reduce the field of view angle, the image is reacquired and verified until it meets the requirements.

[0170] The purpose of this step is to ensure that the camera field of view angle can always produce qualified images through quantitative comparison and dynamic adjustment, providing a high-quality image basis for subsequent edge recognition and calibration.

[0171] Reference Figure 8 The image optimization module is also configured with a light source optimization strategy, including:

[0172] Step S500: feature analysis is performed on the edge image of the bimetal steel strip to determine the edge image shooting area, and the light intensity corresponding to the optimal shooting lens in the preset light database is matched;

[0173] The edge image shooting area refers to the core area where the edge of the bimetal steel strip is located in the image, which is usually the edge and a certain range nearby, to ensure that the edge features are fully presented. The pixel range of this area can be located through feature analysis, such as a 50-pixel band-shaped area on both sides of the edge. The preset light database stores light parameters corresponding to different shooting lenses with different field angles and edge features of the steel strip. The light intensity is a core parameter, for example, a wide field lens needs higher light intensity to cover a larger area, and a narrow field lens can appropriately reduce the intensity to avoid reflection. The light intensity corresponding to the optimal shooting lens refers to the light intensity value that can make the edge of the area and the background form a clear gray difference, facilitating edge recognition.

[0174] In specific implementation, the system locates the edge shooting area from the edge image, matches the corresponding light intensity from the preset light database according to the currently used optimal shooting lens model, such as matching a light intensity of 500 lux for the edge area of a narrow field lens, to provide initial parameters for light source adjustment.

[0175] The purpose of this step is to provide appropriate lighting for the edge area by matching the appropriate light intensity, to avoid edge blur caused by too dark light or reflection caused by too bright light, which affects subsequent feature recognition.

[0176] Step S501: light source adjustment is performed on the edge image shooting area according to the light intensity and the preset parallel illumination angle, and the light source adjustment gain value is calculated by a preset light source gain analysis model, to keep the light source adjustment gain value within the preset reference gain value interval.

[0177] The parallel illumination angle refers to the angle between the light source illumination direction and the surface of the bimetal steel strip, which is preset to 45°, to reduce the specular reflection caused by vertical illumination. The angle is fixed by mechanical structure to ensure uniform illumination of the edge shooting area. The light source adjustment gain value is a parameter used to fine-tune the actual output intensity of the light source, which is calculated based on the initial light intensity, to compensate for the influence of environmental light changes or differences in steel strip surface reflection. The reference gain value interval is the preset normal gain range, such as 0.8-1.2, to ensure that the light source intensity adjustment is within a reasonable range, avoiding excessive adjustment that causes image quality fluctuations.

[0178] In specific implementation, the system first turns on the light source according to the matched light intensity and the preset parallel angle, and then calculates the required adjustment gain value through the light source gain analysis model; if the gain value is within the reference interval, such as 1.0, the light source state is maintained; if the gain value decreases to 0.7 below the lower limit due to environmental light enhancement, the light source intensity is appropriately increased to make the gain value return to the interval.

[0179] The role of this step is to offset the influence of external interference on the light by dynamically adjusting the light source intensity, to ensure that the light of the edge shooting area is always stable, and to provide continuous protection for the clear presentation of the edge features.

[0180] It needs to be further explained that the light source gain analysis model is calculated by the following formula:

[0181]

[0182] Wherein, is the light source adjustment gain value, is the gray scale standard deviation of the adjusted image, for reflecting the uniformity of illumination, is the preset gray scale standard deviation of the image before adjustment, is the edge position detection error, is the preset reference edge error coefficient, is the contrast improvement amount of the adjusted image, is the preset reference contrast.

[0183] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the division of the above functional modules is exemplified, and in actual application, the above functions can be completed by different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. The specific working process of the system, device and unit described above can refer to the corresponding process in the foregoing method embodiments, which will not be repeated here.

[0184] The embodiment of the application provides a computer readable storage medium, which stores a computer program capable of being loaded and executed by a processor.

[0185] The computer storage medium includes, for example, a U disk, a mobile hard disk, a read-only memory (Read-Only Memory, ROM), a random access memory (Random Access Memory, RAM), a magnetic disk or an optical disk, and various media capable of storing program codes.

[0186] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the division of the above functional modules is exemplified, and in actual application, the above functions can be completed by different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. The specific working process of the system, device and unit described above can refer to the corresponding process in the foregoing method embodiments, which will not be repeated here. ​

[0187] The above are only preferred embodiments of the present application and are not intended to limit the protection scope of the present application. Any feature disclosed in the specification, including the abstract and the drawings, can be replaced by other equivalent or similar features unless otherwise stated. That is, each feature is only an example of a series of equivalent or similar features unless otherwise stated.

Claims

1. A dual metal steel strip straightness calibration detection system based on three camera grouping measurement, characterized in that, The utility model relates to a kind of bimetallic strip cutting deviation detection device, including: Camera measurement module, first camera, second camera and third camera are arranged equidistantly along the running path of bimetallic strip after shearing, for respectively collecting the edge image of bimetallic strip in three different positions; Precision control module, with the camera measurement module is connected, is configured with camera precision matching strategy, for adjusting the detection precision of first camera, second camera and third camera edge image acquisition; Image optimization module, with the camera measurement module is connected, is configured with lens distortion calibration strategy, and the lens distortion calibration strategy is used for carrying out distortion correction to camera acquisition image; Image processing module, with the camera measurement module is connected, is configured with deviation adjustment strategy, for calculating and analyzing the straightness deviation value of edge image to determine the shearing deviation of bimetallic strip, and feedback adjustment is carried out; When calculating, the following deviation calculation model formula is used: ； ； ； ； wherein, represents the straightness deviation value of the bimetallic steel strip, , , respectively represent the first position value, the second position value and the third position value, is a position compensation coefficient, is a distortion correction coefficient, is a light source influence coefficient, is a lateral deviation value of the actual position of the second group of cameras from the midpoint of the line connecting the first and third cameras, is a preset reference deviation value, is a preset position influence coefficient, is the difference between the actual deviation value of the image edge pixel after lens distortion correction and the theoretical non-distortion value, is a preset reference distortion pixel value, is a distortion influence coefficient, is the gray scale uniformity of the edge area of the bimetallic steel strip under parallel light source irradiation, is a preset lower limit of the effective uniformity of the light source, is an upper limit of the ideal uniformity of the light source, is a preset light source influence coefficient.

2. The system according to claim 1, wherein, The deviation adjustment strategy includes: According to edge image, feature recognition is carried out to determine the position value corresponding to the edge feature position of bimetallic strip in the field of view of first, second and third cameras, and the position value includes first position value, second position value and third position value; Based on position value and preset deviation calculation model, the straightness deviation value of bimetallic strip is calculated, and compared with preset standard straightness deviation interval to determine deviation result; According to deviation result, the shearing position of bimetallic strip is dynamically feedback adjusted, and the straightness deviation value is kept in preset standard straightness deviation interval.

3. The system according to claim 2, wherein, The camera precision matching strategy includes: According to straightness deviation value, deviation range is determined, and corresponding lens model in preset precision database is matched according to the deviation range, and the lens model has different shooting field angles; According to the corresponding lens model of first camera, second camera and third camera, the optimal shooting lens of the current size specification bimetallic strip is determined, and the shooting lens corresponding to the lens model is switched based on the optimal shooting lens.

4. The system according to claim 1, wherein, The camera measurement module is also configured with camera position pre-calibration strategy, including: According to the motion direction of bimetallic strip, standard coordinate axis orientation is formulated, and first camera and third camera are analyzed according to standard coordinate axis orientation, and the reference line of first camera and second camera coordinates is generated; According to the point-line distance analysis of the corresponding coordinates of second camera located in the middle based on reference line, the error distance of second camera and reference line is determined; Based on error distance, position compensation coefficient is generated by combining preset position compensation model, and the second position value detected by second camera is corrected according to position compensation coefficient.

5. The system according to claim 1, wherein, The lens distortion calibration strategy includes: According to the sub-pixel level corner detection of shooting edge image, the lens distortion parameters of first camera, second camera and third camera are determined; Based on the distortion parameters, a polynomial distortion model is constructed, and the pixel-level distortion compensation value is calculated by inversely mapping the coordinates of the real-time collected bimetallic strip edge image; According to distortion compensation value, camera field angle and edge feature position are dynamically corrected, the compensated straight line measurement error is kept less than preset upper limit deviation value, and the corrected image is smoothed, and edge feature definition is kept.

6. The system according to claim 5, wherein, Also included is a lens field of view enhancement sub-strategy, which optimizes the shooting field of view angle of the camera measurement module by the following steps: According to the shearing parameter matching of the bimetallic steel strip, the corresponding optimal lens field of view angle in the preset database is matched, and the shooting field of view angle of the first camera, the second camera and the third camera is adjusted; The change of the shearing parameter is monitored to determine the trigger field of view angle replacement instruction, the optimal field of view lens angle is updated according to the changed shearing parameter, and the image analysis is carried out according to the updated optimal field of view lens angle with a preset verification sub-strategy, so that the image shooting size of the camera is kept within the preset standard size range.

7. The system according to claim 6, wherein, The verification sub-strategy includes: According to the preset standard image size parameter, the width pixel range, the height pixel range and the edge feature clarity threshold of the optimal image are determined, and the standard image size parameter corresponds to the shearing parameter of the bimetallic steel strip; The image of the bimetallic steel strip under the current field of view angle is collected, and the actual width pixel value, the actual height pixel value and the edge feature clarity value of the image are extracted; Based on the comparison between the actual size value and the standard size parameter, the size deviation value and the clarity deviation value are calculated, if they are all within the preset allowable deviation interval, it is determined that the current field of view angle meets the requirements, if they exceed, the camera field of view angle is fine-tuned according to the deviation value and the image is re-collected until the image size and clarity meet the standard parameter requirements.

8. The system according to claim 3, wherein, The image optimization module is also configured with a light source optimization strategy, including: According to the feature analysis of the edge image of the bimetallic steel strip to determine the edge image shooting area, and matching the light intensity corresponding to the optimal shooting lens in the preset light database; According to the light intensity and the preset parallel illumination angle, the light source of the edge image shooting area is adjusted, and the light source adjustment gain value is calculated by a preset light source gain analysis model, so that the light source adjustment gain value is kept within the preset reference gain value interval.

9. The system according to claim 8, wherein, The light source gain analysis model is calculated by the following formula: ； wherein, is a gain value for the light source, and is a standard deviation of image gray scale after adjustment, for reflecting the uniformity of illumination, is a preset standard deviation of image gray scale before adjustment, is an edge position detection error, is a preset reference edge error coefficient, is a contrast enhancement amount of the image after adjustment, is a preset reference contrast.

Images