H-shaped steel deformation measurement method and system

Abstract

The application provides an H-shaped steel deformation measurement method and system, and belongs to the technical field of profile steel detection.The H-shaped steel deformation measurement system comprises a man-machine interaction module, an image acquisition module, a motion control module and a data processing module.The H-shaped steel deformation measurement method performs profile steel position detection, image acquisition, image processing and deformation parameter calculation through the man-machine interaction module, the image acquisition module, the motion control module and the data processing module, can move the camera to the target position for shooting at the optimal shooting angle, acquires the feature information of the final image through image processing, and thus acquires a higher-precision deformation calculation result, and has comprehensive functions and a wide popularization prospect.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of profile steel detection, and particularly relates to an H-shaped steel deformation measurement method and system. BACKGROUND

[0002] Hot-rolled H-shaped steel has excellent mechanical properties and superior use performance, such as high structural strength, strong anti-seismic capacity, light structure dead weight, large effective use area, etc., and is widely used in the industrial field based on the excellent comprehensive performance of the hot-rolled H-shaped steel. However, in the production process of the H-shaped steel, due to factors such as casting process and storage environment, it is difficult to avoid different degrees of deformation such as flange plate inclination and web eccentricity, which seriously affects the quality of the H-shaped steel product and even causes huge economic losses. Therefore, the deformation measurement technology of the H-shaped steel has a high degree of research value.

[0003] Machine vision technology, as a typical way of non-contact measurement, has become one of the important driving forces for improving production intelligence with the advantages of high precision, strong real-time performance, high automation and intelligence, etc. The machine vision system is a measurement system that obtains the feature image of the detected object by means of optical devices and non-contact sensors, and extracts information from the image and analyzes and processes by the computer, and then realizes detection and control. At present, machine vision technology has been widely applied to the industrial detection field.

[0004] Among them, in the machine vision system, the acquired image is usually processed offline, and the deformation geometric information of the H-shaped steel is obtained from the commonly used features such as corner points, edges, line segments and circles detected, which is the current common analysis method. However, since the production of H-shaped steel is always in a dynamic process, and the above machine vision detection method requires strict shooting position relationship (the camera optical center is strictly perpendicular to the center of the H-shaped steel), it is difficult to obtain accurate measurement results. SUMMARY

[0005] The purpose of the present application is to provide an H-shaped steel deformation measurement method and system, which aims to solve the technical problem that the traditional machine vision detection method is difficult to obtain accurate measurement results.

[0006] To achieve the above purpose, the technical scheme adopted by the present application is:

[0007] In a first aspect, the present application provides an H-shaped steel deformation measurement method, comprising:

[0008] The man-machine interaction module receives the operation command of the user on the graphical interface, and sends the operation command to the data processing module;

[0009] According to the operation instruction, the data processing module sends an execution instruction to the image acquisition module;

[0010] According to the execution instruction, the image acquisition module acquires images in the production process of the profiled steel in real time and sends profiled steel position information to the data processing module;

[0011] According to the profiled steel position information, the data processing module sends an acquisition instruction to the image acquisition module;

[0012] According to the acquisition instruction, the image acquisition module sends an initial image at the current time to the data processing module;

[0013] According to the position and size of the profiled steel in the initial image, the data processing module determines whether the image acquisition module is at a target position, and if not, calculates a motion trajectory of the image acquisition module from a current position to the target position and sends a movement instruction to a motion control module;

[0014] According to the movement instruction and the motion trajectory, the motion control module moves the image acquisition module to the target position;

[0015] The data processing module sends the acquisition instruction to the image acquisition module again;

[0016] According to the acquisition instruction, the image acquisition module sends an initial image at the current time to the data processing module again;

[0017] The data processing module pre-processes the initial image to obtain a final image, calculates deformation information of the profiled steel, and sends the initial image, the final image and the deformation information of the profiled steel to the man-machine interaction module;

[0018] The man-machine interaction module displays the initial image, the final image and the deformation information of the profiled steel on the graphic interface.

[0019] In combination with the first aspect, in some implementations, the data processing module pre-processes the initial image to obtain a final image and calculates deformation information of the profiled steel, including:

[0020] The initial image is corrected to a final image under ideal position conditions by a self-calibration algorithm;

[0021] The final image is used to obtain key line segment information of the profiled steel by using a Gaussian blur algorithm, an edge detection algorithm, a straight line detection algorithm and a line segment feature calculation algorithm;

[0022] According to the key line segment information, the eccentricity and the inclination of the profiled steel are calculated.

[0023] In combination with the first aspect, in some implementations, the initial image is corrected to a final image under ideal position conditions by a self-calibration algorithm, including:

[0024] According to the spatial structure that the flange plates on two sides of the H-shaped steel are parallel to each other and the flange plates on the upper side are parallel to each other, the vanishing points that are orthogonal to each other in the initial image are calculated, the internal and external parameters of the camera in the image acquisition module are calculated, and the final image after camera correction is obtained.

[0025] In combination with the first aspect, in some implementations, the Gaussian blur algorithm adopts a 3*3 operation kernel function, and the weight distribution of the Gaussian kernel function is as follows:

[0026] Wherein, u and v are positions relative to the center (0, 0) of the Gaussian kernel;

[0027] The process of Gaussian blur is represented as:

[0028] Wherein, G(u, v) is the weight of the Gaussian kernel function, X(x+u, y+v) is the pixel value of the original image, and Y(x, y) is the pixel value of the target image.

[0029] In combination with the first aspect, in some implementations, the edge detection algorithm includes: grayscale conversion, filtering and noise reduction, calculation of gradient amplitude and gradient direction, non-maximum suppression, and lag threshold, wherein the calculation formula of the gradient amplitude is as follows:

[0030] Wherein, G x and G y are the gradients in the x direction and the y direction, respectively;

[0031] The calculation formula of the gradient direction is as follows:

[0032] In combination with the first aspect, in some implementations, the calculation formula of the eccentricity is as follows:

[0033] S = |b1-b2| / 2, wherein S is the eccentricity, b1 is the distance from the upper side of the flange plate of the H-shaped steel to the web, and b2 is the distance from the lower side of the flange plate of the H-shaped steel to the web;

[0034] The deviation range allowed by the inclination is as follows:

[0035]

[0036] Wherein, F is the offset distance of the flange plate deformation of the H-shaped steel relative to the flange plate of the normal H-shaped steel, and B is the height of the H-shaped steel.

[0037] The second aspect, the application also provides a kind of H-shaped steel deformation measurement system, including man-machine interaction module, image acquisition module, motion control module and data processing module;

[0038] Wherein,

[0039] The human-computer interaction module is used to: receive operation commands from the user on the graphical interface, send the operation commands to the data processing module, and receive the initial image, final image, and deformation information of the steel section sent by the data processing module and display them on the graphical interface.

[0040] The image acquisition module is used to: receive the execution command sent by the data processing module, acquire images of the steel section production process in real time, send the steel section location information to the data processing module, and send the initial image of the current moment to the data processing module when it receives the acquisition command from the data processing module.

[0041] The motion control module is used to: move the image acquisition module to the target position according to the movement command and motion trajectory sent by the data processing module;

[0042] The data processing module is configured to: send an execution instruction to the image acquisition module according to the operation command; receive the position information of the steel section; send an acquisition instruction to the image acquisition module; receive the initial image at the current moment; determine whether the image acquisition module is at the target position based on the position and size of the steel section in the initial image; if not, calculate the motion trajectory of the image acquisition module from the current position to the target position; send a movement instruction to the motion control module; send an acquisition instruction to the image acquisition module again; receive the initial image at the current moment again; preprocess the initial image to obtain the final image; calculate the deformation information of the steel section; and send the initial image, the final image, and the deformation information of the steel section to the human-computer interaction module.

[0043] In conjunction with the second aspect, in some implementations, the image acquisition module includes:

[0044] The image acquisition unit is used to acquire images in real time during the steel production process and send the initial image at the current moment to the data processing module.

[0045] The position detection unit is used to determine whether the steel section has moved to the preset position and sends the position information of the steel section to the data processing module in real time.

[0046] In conjunction with the second aspect, in some implementations, the motion control module includes:

[0047] A multi-degree-of-freedom robotic arm is used to move the camera of the image acquisition module to the target position;

[0048] An embedded driving unit is used to receive the movement commands and motion trajectories sent by the data processing module, convert the movement commands and motion trajectories into voltage and current signals, detect the joint angle information of the multi-degree-of-freedom robotic arm in real time, and drive the multi-degree-of-freedom robotic arm to move.

[0049] In conjunction with the second aspect, in some implementations, the data processing module includes:

[0050] The instruction sending unit is used to establish a calculation result and instruction transmission path with the human-computer interaction module, the image acquisition module and the motion control module;

[0051] The target position calculation unit is used to plan the motion path of the motion control module, calculate the motion trajectory from the current position to the target position, and send movement commands to the motion control module.

[0052] The self-calibration calculation unit is used to correct the initial image into a final image under ideal position conditions through a self-calibration algorithm;

[0053] A feature extraction unit is used to extract and filter key line segment information from the final image;

[0054] The result calculation unit is used to calculate the eccentricity and inclination of the steel section based on the key line segment information.

[0055] The H-beam deformation measurement method and system provided by this invention have at least the following technical advantages: Compared with traditional technologies, the human-computer interaction module can accept user operation commands and display the initial image, final image, and deformation information of the steel section in real time; the image acquisition module can acquire images of the steel section during production in real time and send the steel section position information and the initial image at the current moment to the data processing module; the motion control module can move the camera in the image acquisition module to the target position according to the movement instructions and trajectory of the data processing module; the data processing module can receive operation commands, send acquisition commands to the image acquisition module multiple times, determine whether the camera is in the target position based on the initial image acquired initially, calculate the movement trajectory of the camera from the current position to the target position based on the steel section position information, send movement commands to the motion control module, preprocess the initial image acquired again to obtain the final image, and calculate the deformation information of the steel section. The above measurement method and system effectively combine the human-computer interaction module, image acquisition module, motion control module, and data processing module. It eliminates the need for cumbersome processes such as steel profile type confirmation and position confirmation, and enables the camera to move to the target position for optimal shooting angle. Through image processing, the feature information of the final image is obtained, thereby obtaining higher precision deformation calculation results. It has comprehensive functions and broad application prospects. Attached Figure Description

[0056] To more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0057] Figure 1 A flowchart of an H-beam deformation measurement method provided in an embodiment of the present invention;

[0058] Figure 2 This is a schematic diagram of the self-calibration correction principle of H-beams in one embodiment of the present invention;

[0059] Figure 3 This is a schematic diagram illustrating the calculation principle of H-beam eccentricity and inclination in one embodiment of the present invention;

[0060] Figure 4 This is a schematic diagram of the frame of an H-beam deformation measurement system provided in an embodiment of the present invention.

[0061] Explanation of reference numerals in the attached figures:

[0062] 100. Human-computer interaction module; 200. Image acquisition module; 210. Image acquisition unit

[0063] 220, Position Detection Unit 300, Motion Control Module 310, Multi-DOF Robotic Arm 320, Embedded Drive Unit 400, Data Processing Module 410, Command Sending Unit

[0064] 420. Target position calculation unit; 430. Self-calibration calculation unit; 440. Result calculation unit Detailed Implementation

[0065] To make the technical problems to be solved, the technical solutions, and the beneficial effects of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present invention and are not intended to limit the present invention.

[0066] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0067] Please refer to the following: Figures 1 to 4 The present invention will now describe the H-beam deformation measurement method and system provided in the embodiments of the present invention.

[0068] Please see Figure 1 and Figure 2This invention provides a method for measuring the deformation of H-beams, comprising:

[0069] S100 The human-computer interaction module receives operation commands from the user on the graphical interface and sends operation commands to the data processing module.

[0070] S101. According to the operation instructions, the data processing module sends the execution instructions to the image acquisition module.

[0071] S102. According to the execution instruction, the image acquisition module acquires images of the steel section production process in real time and sends the steel section position information to the data processing module.

[0072] S103. Based on the position information of the steel profile, the data processing module sends an acquisition command to the image acquisition module.

[0073] S104. According to the acquisition command, the image acquisition module sends the initial image of the current moment to the data processing module.

[0074] S105. Based on the position and size of the steel section in the initial image, the data processing module determines whether the image acquisition module is at the target position. If not, it calculates the motion trajectory of the image acquisition module from the current position to the target position and sends a movement command to the motion control module.

[0075] S106. According to the movement command and motion trajectory, the motion control module moves the image acquisition module to the target position.

[0076] S107. The data processing module sends an acquisition command to the image acquisition module again.

[0077] S108. According to the acquisition command, the image acquisition module sends the initial image of the current moment to the data processing module again.

[0078] S109. The data processing module preprocesses the initial image to obtain the final image, calculates the deformation information of the steel section, and sends the initial image, the final image, and the deformation information of the steel section to the human-computer interaction module.

[0079] S110, the human-computer interaction module displays the initial image, the final image, and the deformation information of the steel section on the graphical interface.

[0080] Specifically, the human-computer interaction module may include a graphical interface and function buttons, or a graphical interface with touch functionality. It can execute operation commands such as initial detection, saving calculation results, stopping initial detection, and starting acquisition operations. Furthermore, it can display the real-time received steel profile images and the final calculation results on the graphical interface, thereby facilitating users to observe and obtain current information.

[0081] The human-computer interaction module can be programmed based on QT 4.9 and compiled using the MinGW 7.3.0-64bit for C++ compiler, or it can be programmed based on other types of C or Java languages, without any restrictions.

[0082] The motion control module includes: a multi-degree-of-freedom robotic arm, used to move the camera of the image acquisition module to the target position; and an embedded drive unit, used to receive the movement commands and motion trajectories sent by the data processing module, convert the movement commands and motion trajectories into voltage and current signals, detect the joint angle information of the multi-degree-of-freedom robotic arm in real time, and drive the multi-degree-of-freedom robotic arm to move.

[0083] Multi-degree-of-freedom (DOF) robotic arms can be five, six, or seven degrees of freedom, etc., without limitation. Through joint movements, the multi-DOF robotic arm can move to a target position, allowing the camera mounted at its end to obtain a suitable shooting angle and field of view. The multi-DOF robotic arm and the camera are fixedly installed using grippers, clamps, snap-fits, or binding methods, without limitation.

[0084] The embedded drive unit consists of hardware such as sensors, motors, and processing chips. It receives movement commands and motion trajectories from the data processing module, converts these commands and trajectories into corresponding voltage and current signals, drives the motor to move, and detects joint angle information in real time, enabling the multi-degree-of-freedom robotic arm to rotate to the target position.

[0085] Joint angle information can be obtained in real time by an angular velocity sensor. The angular velocity sensor detects the joint angular velocity, integrates it over a corresponding time period to obtain the joint angle, and differentiates it over a corresponding time period to obtain the joint acceleration.

[0086] The joint angle information is sent to the data processing module, which can control the start and stop of the embedded drive unit in real time based on the joint angle information, thereby controlling the start and stop of the multi-degree-of-freedom robotic arm.

[0087] The image acquisition module includes: an image acquisition unit, used to acquire images of the steel section production process in real time and send the initial image of the current moment to the data processing module; and a position detection unit, used to determine whether the steel section has moved to the preset position and send the position information of the steel section to the data processing module in real time.

[0088] The image acquisition unit is installed at the end of the multi-degree-of-freedom robotic arm to collect real-time image data during the steel section production process. This data is then transmitted to the data processing module and the human-machine interface module via USB or wireless communication. Specifically, the image acquisition unit is an industrial camera, such as the Daheng Image MER2-U3 Mercury series area array camera. The shooting angle changes with the movement of the multi-degree-of-freedom robotic arm, and the acquired images can be used for graphical interface visualization, multi-degree-of-freedom robotic arm position calculation, and steel section deformation information calculation.

[0089] The position detection unit is installed along the preset movement path of the steel profile to determine its current position. It can return the current position information to the data processing module in real time via USB interface or wireless communication, and can send acquisition commands to the image acquisition unit via software triggering. Specifically, the position detection unit is a position sensor, such as a photoelectric switch, specifically the Omron EE-SX672 series groove-to-beam miniature position sensing photoelectric switch sensor. It is understood that to improve detection accuracy, multiple position sensors can be used simultaneously; these sensors can be of the same or different types.

[0090] The data processing module includes hardware such as a memory and a processor. The memory stores computer programs for executing corresponding functional steps. Any references to memory, storage, databases, or other media used in these embodiments of the invention may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory.

[0091] The H-beam deformation measurement method provided in this invention has at least the following technical advantages: Compared with traditional technologies, the human-computer interaction module can accept user operation commands and display the initial image, final image, and deformation information of the steel section in real time; the image acquisition module can acquire images of the steel section during production in real time and send the steel section position information and the initial image at the current moment to the data processing module; the motion control module can move the camera in the image acquisition module to the target position according to the movement instructions and trajectory of the data processing module; the data processing module can receive operation commands, send acquisition commands to the image acquisition module multiple times, determine whether the camera is in the target position based on the initial image acquired initially, calculate the movement trajectory of the camera from the current position to the target position based on the steel section position information, send movement commands to the motion control module, preprocess the initial image acquired again to obtain the final image, and calculate the deformation information of the steel section. The above measurement method effectively combines the human-computer interaction module, image acquisition module, motion control module, and data processing module. It eliminates the need for cumbersome processes such as steel profile type confirmation and position confirmation, and enables the camera to move to the target position for optimal shooting angle. Through image processing, it obtains the feature information of the final image, thereby obtaining higher precision deformation calculation results. It has comprehensive functions and broad application prospects.

[0092] In some embodiments, before sending the operation command to the data processing module in step S100, system initialization steps such as initialization detection and stopping initialization detection are also included.

[0093] In some embodiments, in steps S100-S110, the image acquisition module can send the acquired steel profile image directly to the human-computer interaction module or to the human-computer interaction module through the data processing module in real time, or send the video stream of the entire shooting process, so that the human-computer interaction module can display the steel profile acquisition status in real time.

[0094] In some embodiments, in step S105, the data processing module can obtain a binary image of the outer contour of the steel profile through image adaptive threshold segmentation, and calculate the center position and area of ​​the steel profile through connected component analysis to determine the approximate position and angle of the steel profile from the camera. Then, by calculating the difference between the center position of the profile and the center position of the image, and the difference between the area of ​​the profile and the area of ​​the image, the motion path of the multi-degree-of-freedom robotic arm is planned, that is, the motion trajectory from the current position to the target position is calculated, and the calculation result is sent to the embedded drive unit to control the multi-degree-of-freedom robotic arm to move the camera to the target position.

[0095] In some embodiments, the data processing module preprocesses the initial image to obtain the final image and calculates the deformation information of the steel profile, including: correcting the initial image to the final image under ideal position conditions using a self-calibration algorithm; using a Gaussian blur algorithm, an edge detection algorithm, a straight line detection algorithm, and a line segment feature statistical algorithm on the final image to obtain the key line segment information of the steel profile; and calculating the eccentricity and tilt of the steel profile based on the key line segment information.

[0096] The data processing module described above employs a variety of comprehensive processing algorithms to process the initial image. Through repeated experiments, more accurate key line segment information of the H-beam can be obtained, thereby effectively acquiring the outer contour information of the H-beam and ensuring accurate measurement of eccentricity and tilt.

[0097] Combining the above image processing methods, such as Figure 3 As shown, in some embodiments, the initial image is corrected to a final image under ideal positioning conditions using a self-calibration algorithm. This includes: calculating the mutually orthogonal vanishing points in the initial image based on the spatial structure of the parallel flanges on both sides of the H-beam and the parallel upper flange; calculating the camera's intrinsic and extrinsic parameters in the image acquisition module; and obtaining the final image after camera correction. This setup, by solving for the extrinsic parameters to obtain the corrected image, avoids errors caused by the relative rotation angle and translation of the H-beam and the camera's optical center, thus reducing the stringent requirements on the relative position of the camera's optical center and the H-beam during dynamic production.

[0098] Specifically, firstly, the camera shooting process is equivalent to transforming a three-dimensional Euclidean space into a two-dimensional projective image through a projective transformation. This process can be expressed as a formula:

[0099]

[0100] Among them, (u i v i (x) represents the image coordinates. i y i , z i () represents the coordinates in the world coordinate system.

[0101] Second, due to P 3×4 Since it is a transformation matrix, it can be decomposed into:

[0102] P = K[RT] (0.2)

[0103] The matrix composed of R and T is the extrinsic parameter matrix, which is the rotation and translation matrix of the camera coordinate system relative to the world coordinate system.

[0104] Third, matrix K is called the intrinsic parameter matrix, which consists of the camera focal length, pixel ratio, and the horizontal and vertical translations of the camera optical center relative to the origin. The formula is:

[0105]

[0106] Fourth, the edge segments of the steel flange are obtained from the acquired images according to the preset image processing steps. Key points and line segment feature information are obtained from the edge segments of the steel flange. Based on the key points and line segment feature information, the vanishing points p1, p2, and p3, which are parallel lines on the steel section and parallel lines on the flange that are orthogonal to each other, are calculated. The coordinates are represented as (u p1 v p1 ), (u p2 v p2 ) and (u p3 v p3 Then, the orthocenter of Δp1p2p3, i.e., the projection center p(u), can be calculated. o v o Therefore, the coordinates (x, y) of p1 in the camera coordinate system can be obtained. c1 y c1 f):

[0107]

[0108] Similarly, the coordinates (x, y) of p2 in the camera coordinate system can be obtained. c2 y c2 f):

[0109]

[0110] Based on the properties of the vanishing point, the focal length intrinsic parameter f can be obtained, and the calculation formula is as follows:

[0111]

[0112] Fifth, the formula for calculating the camera extrinsic parameter matrix R is:

[0113]

[0114] Sixth, the formula for calculating the translation matrix T is:

[0115]

[0116] in, The calculation formula is:

[0117]

[0118] Based on the calculation formulas (0.2) to (0.9) and the corresponding point relationships, all parameters inside and outside the camera can be calculated. Finally, the perspective transformation matrix can be obtained from the results of the above calculation formulas, at which point the image can be corrected.

[0119] In conjunction with the above image processing methods, in some embodiments, due to the complex environment of industrial production sites, some speckles caused by external impulse noise are inevitably generated during image acquisition. Therefore, an appropriate Gaussian blur algorithm is used to process the noise. Furthermore, to avoid the problem of subsequent operations being difficult to measure due to excessive Gaussian blur, a 3×3 kernel function can be used. The weight distribution of the Gaussian kernel function is as follows:

[0120] Where u and v are the positions relative to the center of the Gaussian kernel (0, 0), respectively.

[0121] Based on the Gaussian kernel function weight distribution formula, the Gaussian blurring process can be represented as follows:

[0122] Where G(u, v) is the weight of the Gaussian kernel function, X(x+u, y+v) is the pixel value of the original image, and Y(x, y) is the pixel value of the target image.

[0123] In combination with the above image processing methods, in some embodiments, the edge detection algorithm needs to go through a series of image processing methods, specifically: grayscale image conversion, filtering and noise reduction, calculation of gradient magnitude and direction, non-maximum suppression and hysteresis threshold, so as to obtain the edge calculation result.

[0124] In this embodiment, the formula for calculating the gradient magnitude is:

[0125] Among them, G x and G y These are the gradients in the x and y directions, respectively;

[0126] The formula for calculating the gradient direction is:

[0127] In this embodiment, Non-Maximum Suppression (NMS) suppresses elements that are not local maxima. In simple terms, it preserves the maximum gradient intensity at each pixel while deleting other values, which helps suppress all gradient values ​​except for local maxima.

[0128] In this embodiment, hysteresis thresholding requires two thresholds: a high threshold and a low threshold. It is assumed that significant edges in the image are continuous curves, allowing for the tracking of blurred portions of a given curve and avoiding the misidentification of noise pixels that do not constitute a curve as edges. Therefore, a larger threshold is started, which identifies more certain true edges. Using the previously derived direction information, the entire edge in the image is tracked from these true edges. During tracking, a smaller threshold is used, allowing for the tracking of blurred portions of the curve until returning to the starting point. Through repeated experiments based on the H-beam steel image acquisition environment, a 3:1 ratio of the maximum to the minimum threshold yielded the best results in calculating the hysteresis threshold.

[0129] In combination with the above image processing methods, in some embodiments, the line detection algorithm can detect a set of pixels that conform to the characteristics of a line through a voting algorithm. This process obtains a set that conforms to the specific shape by calculating the local maximum value of the cumulative result in the parameter space, which is then used as the result of the Hough transform.

[0130] In combination with the above image processing methods, in some embodiments, the line segment feature statistics method filters the above line segment feature information based on the geometric characteristics of H-beams and less prior knowledge. For example, it filters layer by layer from line segment length, line segment angle, and line segment position distribution concentration to finally obtain more accurate information on the outer edge of H-beams.

[0131] Combining the above image processing methods, such as Figure 4 As shown, in some embodiments, the formula for calculating eccentricity is:

[0132] S = |b1-b2| / 2, where S is the eccentricity, b1 is the distance from the upper side of the flange plate to the web of the H-beam, and b2 is the distance from the lower side of the flange plate to the web of the H-beam.

[0133] According to the national standard GB / T 11263-2017 "Hot-rolled H-beams and some T-beams", the tilt angle is determined to be within the allowable deviation range of the finished product. The allowable deviation range of the tilt angle is as follows:

[0134]

[0135] Where F is the offset distance of the flange of the H-beam relative to the flange of the normal H-beam after deformation, and B is the height of the H-beam.

[0136] Through the above formula, due to the rigorous calculation and screening of key line segment information in the early stage, more accurate outer contour information is obtained, so the eccentricity and inclination of the steel section can be accurately calculated.

[0137] The following are device embodiments of this application. For details not described in detail, please refer to the corresponding method embodiments described above. It should be assumed that the details not described in detail in the device embodiments are the same as those in the method embodiments and have been clearly recorded in the specification.

[0138] Based on the same inventive concept, please refer to Figure 2 This invention also provides an H-beam deformation measurement system, including a human-computer interaction module, an image acquisition module, a motion control module, and a data processing module.

[0139] Specifically, the human-computer interaction module is used to: receive operation commands from the user on the graphical interface, send operation commands to the data processing module, and receive and display the initial image, final image, and deformation information of the steel section sent by the data processing module on the graphical interface.

[0140] The image acquisition module is used to: receive execution instructions sent by the data processing module, acquire images of the steel profile production process in real time, send the steel profile position information to the data processing module, and send the initial image of the current moment to the data processing module when it receives the acquisition command from the data processing module.

[0141] The motion control module is used to move the image acquisition module to the target position according to the movement instructions and motion trajectory sent by the data processing module.

[0142] The data processing module is used to: send execution instructions to the image acquisition module according to the operation command, receive the position information of the steel section, send acquisition instructions to the image acquisition module, receive the initial image at the current moment, determine whether the image acquisition module is at the target position based on the position and size of the steel section in the initial image, if not, calculate the motion trajectory of the image acquisition module from the current position to the target position, send a movement instruction to the motion control module, send an acquisition instruction to the image acquisition module again, receive the initial image at the current moment again, preprocess the initial image to obtain the final image, calculate the deformation information of the steel section, and send the initial image, the final image, and the deformation information of the steel section to the human-computer interaction module.

[0143] The H-beam deformation measurement system provided in this embodiment of the invention has the same technical effect as the above-described H-beam deformation measurement method, and will not be repeated here.

[0144] The description of the human-computer interaction module in this embodiment of the invention is consistent with that in the above method, and will not be repeated here.

[0145] In some embodiments, the image acquisition module includes: an image acquisition unit, used to acquire images of the steel section production process in real time and send the initial image at the current moment to the data processing module; and a position detection unit, used to determine whether the steel section has moved to a preset position and send the position information of the steel section to the data processing module in real time. The description of the image acquisition module in this embodiment is consistent with that in the above method and will not be repeated here.

[0146] In some embodiments, the motion control module includes: a multi-degree-of-freedom robotic arm for moving the camera of the image acquisition module to a target position; and an embedded drive unit for receiving movement commands and motion trajectories sent by the data processing module, converting the movement commands and motion trajectories into voltage and current signals, detecting the joint angle information of the multi-degree-of-freedom robotic arm in real time, and driving the multi-degree-of-freedom robotic arm to move. The description of the motion control module in this embodiment is consistent with that in the above method and will not be repeated here.

[0147] In some embodiments, the data processing module includes: an instruction sending unit for establishing a calculation result and instruction transmission path with the human-computer interaction module, the image acquisition module, and the motion control module; a target position calculation unit for planning the motion path of the motion control module, calculating the motion trajectory from the current position to the target position, and sending a movement instruction to the motion control module; a self-calibration calculation unit for correcting the initial image to the final image under ideal position conditions through a self-calibration algorithm; a feature extraction unit for extracting and filtering key line segment information from the final image; and a result calculation unit for calculating the eccentricity and inclination of the steel profile based on the key line segment information.

[0148] Specifically, the target position calculation unit can obtain a binary image of the outer contour of the steel profile through image adaptive threshold segmentation, and calculate the center position and area of ​​the steel profile through connected component analysis to determine the approximate position and angle of the steel profile from the camera. Then, by calculating the difference between the center position of the profile and the center position of the image, and the difference between the area of ​​the profile and the area of ​​the image, the motion path of the multi-degree-of-freedom robotic arm is planned, that is, the motion trajectory from the current position to the target position is calculated, and the calculation results are sent to the command sending unit, which sends them to the embedded drive unit, thereby controlling the multi-degree-of-freedom robotic arm to move the camera to the target position.

[0149] Specifically, based on the characteristic that most H-shaped steel structures are composed of straight line segments, the feature extraction unit uses Gaussian blur algorithm, edge detection algorithm, straight line detection algorithm and line segment feature statistics method to obtain key line segment information for calculating deformation parameters from the corrected image.

[0150] Specifically, the result calculation unit can calculate the eccentricity and tilt according to the national standard GB / T 11263-2017 "Hot-rolled H-beams and some T-beams" and determine whether they are within the allowable deviation of the finished product.

[0151] It should be noted that the human-computer interaction module, image acquisition module, motion control module, and data processing module provided in the embodiments of the present invention are used to achieve the same functions as those in the above methods, and will not be described again here.

[0152] The module division in this embodiment of the invention is illustrative and represents only one logical functional division; in actual implementation, other division methods may be used. Each module in the above-described device can be implemented entirely or partially through software, hardware, or a combination thereof. Each module can be embedded in hardware within or independently of the processor in a computer device, or stored in software in the memory of a computer device, so that the processor can call and execute the operations corresponding to each module.

[0153] It is understood that the parts in the above embodiments can be freely combined or deleted to form different combined embodiments. The specific contents of each combined embodiment will not be repeated here. After this description, it can be considered that the specification has recorded each combined embodiment and can support different combined embodiments.

[0154] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for measuring the deformation of H-beams, characterized in that, include: The human-computer interaction module receives operation commands from the user on the graphical interface and sends the operation commands to the data processing module. According to the operation instructions, the data processing module sends an execution instruction to the image acquisition module; According to the execution instruction, the image acquisition module acquires images of the steel section production process in real time and sends the steel section location information to the data processing module; Based on the steel section location information, the data processing module sends an acquisition command to the image acquisition module; According to the acquisition command, the image acquisition module sends the initial image of the current moment to the data processing module; Based on the position and size of the steel section in the initial image, the data processing module determines whether the image acquisition module is at the target position. If not, it calculates the motion trajectory of the image acquisition module from the current position to the target position and sends a movement command to the motion control module. According to the movement command and the movement trajectory, the motion control module moves the image acquisition module to the target position; The data processing module sends the acquisition command to the image acquisition module again; According to the acquisition command, the image acquisition module sends the initial image of the current moment to the data processing module again; The data processing module preprocesses the initial image acquired at the target location to obtain the final image, calculates the deformation information of the steel section, and sends the initial image, the final image, and the deformation information of the steel section to the human-computer interaction module. The human-computer interaction module displays the initial image, the final image, and the deformation information of the steel section on the graphical interface.

2. The method according to claim 1, characterized in that, The data processing module preprocesses the initial image acquired at the target location to obtain the final image, and calculates the deformation information of the steel section, including: The initial image is corrected to a final image under ideal positioning conditions using a self-calibration algorithm; The key line segment information of the steel profile is obtained by using Gaussian blur algorithm, edge detection algorithm, line detection algorithm and line segment feature statistics algorithm on the final image; Based on the key line segment information, calculate the eccentricity and inclination of the steel section.

3. The method according to claim 2, characterized in that, The initial image is corrected to a final image under ideal positioning conditions using a self-calibration algorithm, including: Based on the spatial structure of the parallel flanges on both sides of the H-beam and the parallel upper flange, calculate the mutually orthogonal vanishing points in the initial image, calculate the camera's intrinsic and extrinsic parameters in the image acquisition module, and obtain the final image after camera correction.

4. The method according to claim 2, characterized in that, The Gaussian blur algorithm uses a 3×3 kernel function, and the weight distribution of the Gaussian kernel function is as follows: ,in, u and v These are the positions relative to the center of the Gaussian kernel (0, 0); The process of Gaussian blurring is represented as: ,in, The weights are Gaussian kernel function weights. These are the pixel values ​​of the original image. The pixel value of the target image.

5. The method according to claim 2, characterized in that, The edge detection algorithm includes: grayscale image conversion, filtering and noise reduction, calculation of gradient magnitude and gradient direction, non-maximum suppression, and hysteresis thresholding, wherein the formula for calculating the gradient magnitude is: ,in, and They are respectively x direction and y Gradient in direction; The formula for calculating the gradient direction is: .

6. The method according to claim 2, characterized in that, The formula for calculating the eccentricity is: Where S is the eccentricity, b1 is the distance from the upper side of the flange plate to the web of the H-beam, and b2 is the distance from the lower side of the flange plate to the web of the H-beam. The allowable deviation range for the tilt angle is: ， Where F is the offset distance of the flange of the H-beam relative to the flange of the normal H-beam after deformation, and B is the height of the H-beam.

7. A deformation measurement system for H-beams, characterized in that, It includes a human-computer interaction module, an image acquisition module, a motion control module, and a data processing module; in, The human-computer interaction module is used to: receive operation commands from the user on the graphical interface, send the operation commands to the data processing module, and receive the initial image, final image, and deformation information of the steel section sent by the data processing module and display them on the graphical interface. The image acquisition module is used to: receive the execution command sent by the data processing module, acquire images of the steel section production process in real time, send the steel section location information to the data processing module, and send the initial image of the current moment to the data processing module when it receives the acquisition command from the data processing module. The motion control module is used to: move the image acquisition module to the target position according to the movement command and motion trajectory sent by the data processing module; The data processing module is configured to: send an execution instruction to the image acquisition module according to the operation command; receive the position information of the steel section; send an acquisition instruction to the image acquisition module; receive the initial image at the current moment; determine whether the image acquisition module is at the target position based on the position and size of the steel section in the initial image; if not, calculate the motion trajectory of the image acquisition module from the current position to the target position; send a movement instruction to the motion control module; send an acquisition instruction to the image acquisition module again; receive the initial image at the current moment again; preprocess the initial image acquired at the target position to obtain the final image; calculate the deformation information of the steel section; and send the initial image, the final image, and the deformation information of the steel section to the human-computer interaction module.

8. The system according to claim 7, characterized in that, The image acquisition module includes: The image acquisition unit is used to acquire images in real time during the steel production process and send the initial image at the current moment to the data processing module. The position detection unit is used to determine whether the steel section has moved to the preset position and sends the position information of the steel section to the data processing module in real time.

9. The system according to claim 7, characterized in that, The motion control module includes: A multi-degree-of-freedom robotic arm is used to move the camera of the image acquisition module to the target position; An embedded driving unit is used to receive the movement commands and motion trajectories sent by the data processing module, convert the movement commands and motion trajectories into voltage and current signals, detect the joint angle information of the multi-degree-of-freedom robotic arm in real time, and drive the multi-degree-of-freedom robotic arm to move.

10. The system according to claim 7, characterized in that, The data processing module includes: The instruction sending unit is used to establish a calculation result and instruction transmission path with the human-computer interaction module, the image acquisition module and the motion control module; The target position calculation unit is used to plan the motion path of the motion control module, calculate the motion trajectory from the current position to the target position, and send movement commands to the motion control module. The self-calibration calculation unit is used to correct the initial image into a final image under ideal position conditions through a self-calibration algorithm; A feature extraction unit is used to extract and filter key line segment information from the final image; The result calculation unit is used to calculate the eccentricity and inclination of the steel section based on the key line segment information.

Images