CN117422664A - Processing and identifying unit, system and method for three-dimensional detection image of hot rolled strip steel edge wave - Google Patents
Processing and identifying unit, system and method for three-dimensional detection image of hot rolled strip steel edge wave Download PDFInfo
- Publication number
- CN117422664A CN117422664A CN202210804607.XA CN202210804607A CN117422664A CN 117422664 A CN117422664 A CN 117422664A CN 202210804607 A CN202210804607 A CN 202210804607A CN 117422664 A CN117422664 A CN 117422664A
- Authority
- CN
- China
- Prior art keywords
- image
- strip steel
- dimensional
- edge
- laser
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 229910000831 Steel Inorganic materials 0.000 title claims abstract description 104
- 239000010959 steel Substances 0.000 title claims abstract description 104
- 238000001514 detection method Methods 0.000 title claims abstract description 86
- 238000012545 processing Methods 0.000 title claims abstract description 84
- 238000000034 method Methods 0.000 title claims abstract description 39
- 230000007547 defect Effects 0.000 claims abstract description 28
- 238000000605 extraction Methods 0.000 claims abstract description 24
- 238000004519 manufacturing process Methods 0.000 claims abstract description 18
- 238000004364 calculation method Methods 0.000 claims abstract description 15
- 238000001914 filtration Methods 0.000 claims abstract description 14
- 230000009467 reduction Effects 0.000 claims abstract description 10
- 230000009466 transformation Effects 0.000 claims abstract description 10
- 238000012937 correction Methods 0.000 claims abstract description 9
- 238000010586 diagram Methods 0.000 claims abstract description 9
- 238000004088 simulation Methods 0.000 claims abstract description 7
- 238000004422 calculation algorithm Methods 0.000 claims description 23
- 238000003860 storage Methods 0.000 claims description 22
- 230000005484 gravity Effects 0.000 claims description 20
- 230000005540 biological transmission Effects 0.000 claims description 14
- 230000008569 process Effects 0.000 claims description 14
- 239000000284 extract Substances 0.000 claims description 12
- 230000003287 optical effect Effects 0.000 claims description 12
- 230000001186 cumulative effect Effects 0.000 claims description 6
- 238000009826 distribution Methods 0.000 claims description 6
- 238000005315 distribution function Methods 0.000 claims description 6
- 238000013507 mapping Methods 0.000 claims description 6
- 239000011159 matrix material Substances 0.000 claims description 5
- 238000004148 unit process Methods 0.000 claims description 5
- 238000004891 communication Methods 0.000 claims description 4
- 230000001681 protective effect Effects 0.000 claims description 2
- 238000007689 inspection Methods 0.000 claims 2
- 238000003384 imaging method Methods 0.000 description 6
- 238000013461 design Methods 0.000 description 4
- 238000005098 hot rolling Methods 0.000 description 3
- 230000033001 locomotion Effects 0.000 description 3
- 230000002159 abnormal effect Effects 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 238000005096 rolling process Methods 0.000 description 2
- 230000011218 segmentation Effects 0.000 description 2
- 241000226585 Antennaria plantaginifolia Species 0.000 description 1
- 241000277275 Oncorhynchus mykiss Species 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000001816 cooling Methods 0.000 description 1
- 230000002950 deficient Effects 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000007730 finishing process Methods 0.000 description 1
- 238000007667 floating Methods 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000009182 swimming Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Chemical compound O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
- 238000003466 welding Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
- G06T7/0008—Industrial image inspection checking presence/absence
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B11/00—Measuring arrangements characterised by the use of optical techniques
- G01B11/24—Measuring arrangements characterised by the use of optical techniques for measuring contours or curvatures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/20—Image enhancement or restoration using local operators
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/40—Image enhancement or restoration using histogram techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30136—Metal
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Geometry (AREA)
- Quality & Reliability (AREA)
- Computer Graphics (AREA)
- Software Systems (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
- Length Measuring Devices By Optical Means (AREA)
Abstract
The invention discloses a processing and identifying unit, a system and a method for three-dimensional detection image of hot rolled strip steel edge waves, comprising the following steps: the image enhancement module is used for carrying out noise reduction and enhancement treatment on the original image on the surface of the strip steel; the jitter judging and removing module is used for judging jitter phenomenon generated by field production overload; the laser line extraction module is used for obtaining the accurate position of the laser line in the image; the laser line correction module is used for carrying out inverse transformation on the laser lines in the image so as to obtain space contour data; the three-dimensional morphology reconstruction module is used for carrying out vibration filtering on the spatial profile data, then merging the spatial profile data, drawing the spatial profile data in a three-dimensional coordinate system to obtain a three-dimensional lattice diagram of the surface of the strip steel, and finally obtaining three-dimensional graphic information of the surface of the whole strip steel through calculation and simulation; and the three-dimensional defect detection and identification module is used for judging whether the strip steel edge wave defect exists or not from the three-dimensional graphic information and determining the defect position. The invention realizes the calculation of the wave distance and the wave height of the side waves of the strip steel, thereby realizing the accurate detection of the side waves of the strip steel.
Description
Technical Field
The invention relates to a strip steel surface visual detection technology, in particular to a hot rolled strip steel edge wave three-dimensional detection image processing and identifying unit, system and method.
Background
When certain specific varieties are produced by hot rolling, the outstanding quality problems are that the head and tail of the strip steel are deviated and single-side waves are defective, the stability of the welding and rolling in the subsequent working procedures is affected, and abnormal shutdown treatment is caused. Factors that produce and affect the shape of hot rolled strip are: the thickness and width of the slab itself, the slab temperature, the roll levelness, etc. They often appear as side waves and camber.
At present, a flatness meter is arranged at the outlet of part of the hot rolling production line, and the flatness meter can be used for detecting flatness conditions of a finish rolling frame of strip steel according to temperature, stress and other graphs. Because the running states of the strip steel head frame and the strip steel tail frame are unstable, phenomena such as swimming, floating and the like of different degrees exist, and after the strip steel passes through a laminar cooling area, the actual plate shape and the high temperature condition have larger changes, so that the detection precision of the flatness meter is low, and the abnormal plate shape influencing the production of the subsequent process cannot be effectively identified. The downstream quality problem caused by the side wave problem in the prior hot rolling mill is quite large.
In the prior patent application, as in patent CN103486995a, three sets of distance meters are used to realize the detection of the plate shape of the plate. Patent CN104833317A discloses a medium plate morphology detection system and method based on controllable symmetrical double-line laser angles. In the invention, double-line laser and two cameras are adopted to realize plate shape detection, a Vision module of Labview is used for directly reading the obtained picture information, processing and splicing the pictures, and finally, coordinate extraction at two ends of a bright line is carried out, so that longitudinal coordinates are obtained, and the whole profile information of the detected steel plate is finally obtained in a curve fitting mode.
The above patent basically adopts a method of combining machine vision with laser lines, but does not realize accurate detection of the edge wave shape of the strip steel.
Disclosure of Invention
Aiming at the defects in the prior art, the invention aims to provide a processing and identifying unit, a system and a method for three-dimensional detection image of hot rolled strip steel edge waves, which are used for eliminating shaking repeated lines in images, realizing calculation of the wave distance and the wave height of strip steel edge waves and further realizing accurate detection of strip steel edge waves.
In order to achieve the above purpose, the invention adopts the following technical scheme:
the first aspect of the invention provides a hot rolled strip edge wave three-dimensional detection image processing and identifying unit, which comprises:
the image enhancement module is used for carrying out noise reduction and enhancement treatment on the original image on the surface of the strip steel;
the jitter judging and removing module is used for judging jitter phenomenon generated by field production overload;
the laser line extraction module is used for obtaining the accurate position of the laser line in the image;
the laser line correction module is used for carrying out inverse transformation on the laser lines in the image so as to obtain space contour data;
the three-dimensional morphology reconstruction module is used for carrying out vibration filtering on the spatial profile data, then merging the spatial profile data, drawing the spatial profile data in a three-dimensional coordinate system to obtain a three-dimensional lattice diagram of the surface of the strip steel, and finally obtaining three-dimensional graphic information of the surface of the whole strip steel through calculation and simulation;
and the three-dimensional defect detection and identification module is used for judging whether the strip steel edge wave defect exists or not from the three-dimensional graphic information and determining the defect position.
Preferably, the image enhancement module performs noise reduction and enhancement processing on the original image by using a Gaussian filtering algorithm and a histogram equalization image processing algorithm;
the Gaussian filtering algorithm adopts a 5×5 Gaussian template, and the specific processing procedure is as follows:
1) Firstly calculating a Gaussian filter kernel of the Gaussian template according to the probability of standard two-dimensional normal distribution;
2) Coinciding each pixel point of the original image with a center point of the Gaussian filter kernel;
3) Multiplying and summing the original image and two 5x5 matrix corresponding elements overlapped by the Gaussian filter kernel, dividing the sum by a value obtained by normalizing parameter 273, and replacing the value of the original image position;
the specific processing procedure of the histogram equalization image processing algorithm is as follows:
1) Scanning each pixel of the original image gray level in turn, and calculating a gray level histogram of the image;
2) Calculating a cumulative distribution function of the gray level histogram;
3) Obtaining a mapping relation between input and output according to the cumulative distribution function and the histogram equalization principle;
4) And obtaining a result according to the mapping relation and carrying out image transformation.
Preferably, the shake judging and removing module is used for removing shake repeated lines existing in the image through image processing. The specific method comprises the following steps: in practical design, a fixed number of overlapping laser lines is set for the front and rear images. If the number of the laser lines which are overlapped by the front image and the rear image is increased or reduced (combined with the judgment of the background environment of the picture) in the detection process through image comparison, the strip steel is considered to have jitter in the motion process. And eliminating repeated laser line images, and avoiding error calculation of defect sizes and positions.
Preferably, the laser line extraction module adopts a gray-scale gravity center laser line extraction algorithm to accurately extract the laser line;
the gray level gravity center laser line extraction algorithm is to process the gray level distribution characteristics in the cross section of each line of light stripe line by line, calculate and extract the gray level gravity center point of the light stripe area line by line in the direction of line coordinates, use the gray level gravity center point to represent the center point position of the light stripe of the cross section, and finally fit all the center points to form the center line of the light stripe;
the gray center of the object S in the gray image I (I, j) is (x) 0 ，y 0 ) The method comprises the following steps:
in the scheme, the specific operation steps are that filtering is firstly carried out on an image to remove noise, then threshold segmentation is carried out, a part larger than the threshold is reserved, namely an ROI (region of interest) area, and then a gray-scale gravity center method is used for extracting a laser center line in the area.
Preferably, the gray-scale gravity center laser line extraction algorithm firstly adopts the maximum value to initially extract the center of the laser line, and then realizes the extraction precision of the laser line through weighting the gray-scale gravity center.
Preferably, the line width pixel number W of the laser line is more than or equal to 8.
The second aspect of the invention provides a hot-rolled strip edge wave three-dimensional detection image processing and identifying system, which comprises a detection triggering unit, a laser structure light irradiator, a high-speed camera, an optical system view field adjusting unit, an image acquisition/transmission unit, an image storage/display unit and the hot-rolled strip edge wave three-dimensional detection image processing and identifying unit provided by the first aspect of the invention;
the detection triggering unit is used for capturing the signal of the strip steel on the roller way and triggering the starting of the laser structure light irradiator and the high-speed camera;
the laser structure light irradiator is arranged above the strip steel and emits laser rays to illuminate the surface of the strip steel;
the high-speed camera is also arranged above the strip steel and is used for shooting an edge surface image of the strip steel;
the optical system view field adjusting unit is used for adjusting the view field of the high-speed camera;
the image acquisition/transmission unit is used for acquiring the edge surface image of the high-speed camera and transmitting the edge surface image to the hot-rolled strip steel edge wave three-dimensional detection image processing and identifying unit;
the hot rolled strip steel edge wave three-dimensional detection image processing and identifying unit processes the edge surface image, calculates and identifies the wave distance and the height of the strip steel upper edge wave, and transmits the wave distance and the height to the image storage/display unit;
the image storage/display unit is used for realizing storage and alarm of the edge surface image.
Preferably, the detection triggering unit comprises a photoelectric correlation switch and an encoder which are arranged on the roller way;
the number N of the laser lines is more than or equal to 2.
The minimum exposure time t=h/v of the high-speed camera, wherein h is v which is the production speed of the belt steel;
the image acquisition/transmission unit is an image acquisition card;
the hot-rolled strip steel edge wave three-dimensional detection image processing and identifying unit is an image processing computer;
the image storage/display unit includes a data server and a terminal computer in communication therewith.
Preferably, the laser structure light irradiator, the high-speed camera and the optical system view field adjusting unit are all arranged in the protective box.
The third aspect of the invention provides a hot-rolled strip edge wave three-dimensional detection image processing and identifying method, adopting the hot-rolled strip edge wave three-dimensional detection image processing and identifying system provided by the second aspect of the invention, capturing the strip passing signal through the detection triggering unit, triggering the laser structure light irradiator and the high-speed camera to start, the laser structure light irradiator sends N parallel equidistant laser lines which are perpendicular to the strip, the optical system view field adjusting unit firstly adjusts the angle theta between the laser lines and the view field of the high-speed camera, the high-speed camera shoots the edge surface image of the strip, the image acquisition/transmission unit is used for acquiring the edge surface image and transmitting the edge surface image to the hot-rolled strip edge wave three-dimensional detection image processing and identifying unit, the hot-rolled strip edge wave three-dimensional detection image processing and identifying unit processes the edge surface image, calculates and identifies the edge wave distance and the height of the strip, and transmits the edge wave distance to the image storage/display unit, and the image storage/display unit is used for realizing the storage and alarm of the image;
the hot-rolled strip steel edge wave three-dimensional detection image processing and identifying unit processes the edge surface image as follows:
the image enhancement module performs noise reduction and enhancement processing on the edge surface image, the jitter judgment and removal module performs jitter judgment on the edge surface image and removes the existing jitter phenomenon, the laser line extraction module extracts laser lines in the edge surface image, the laser line correction module eliminates distortion in the edge surface image and performs inverse transformation on the laser lines so as to obtain space contour data, the three-dimensional shape reconstruction module performs vibration filtering on the space contour data and then merges the space contour data, the space contour data is drawn in a three-dimensional coordinate system to obtain a three-dimensional lattice diagram of the strip steel surface, finally three-dimensional graph information of the whole strip steel surface is obtained through calculation and simulation, the three-dimensional defect detection recognition module judges whether the strip steel edge wave defect exists or not from the three-dimensional graph information and determines the defect position.
According to the processing and identifying unit, system and method for the three-dimensional detection image of the hot rolled strip steel edge wave, in the field use process, the laser structure light source is made into parallel light beams with different laser lines according to the field detection area and field requirement. The parallel light beam is precisely calibrated and adjusted to form the laser linear array light source with higher parallel precision. The optical system view field adjusting unit is used for accurately adjusting and setting the included angle theta between the view field of the high-speed camera and the light rays of the laser structure. The laser structure light source irradiates on the surface of the edge of the strip steel, if the strip steel has edge waves, the laser line on the surface of the strip steel can deviate, a high-speed camera can continuously shoot a plurality of stripes with wave troughs or wave crests, an image acquisition/transmission unit sends the acquired image of the edge of the strip steel to a hot-rolled strip steel edge wave three-dimensional detection image processing and identifying unit, and the hot-rolled strip steel edge wave three-dimensional detection image processing and identifying unit continuously splices, processes and analyzes the image, so that the calculation of the wave distance and the wave height of the edge wave shape of the strip steel can be realized, and the accurate detection of the edge wave shape of the strip steel is realized.
Drawings
FIG. 1 is a schematic diagram of a frame of a system for processing and recognizing three-dimensional detection images of edge waves of hot rolled strip steel;
FIG. 2 is a schematic diagram of the processing flow of the processing and identifying unit for the three-dimensional detection image of the edge wave of the hot rolled strip steel.
Detailed Description
In order to better understand the above technical solution of the present invention, the technical solution of the present invention is further described below with reference to the accompanying drawings and examples.
Referring to fig. 1, the invention provides a three-dimensional detection image processing and identifying unit for hot rolled strip edge waves, which comprises:
the field view environment of the image enhancement module 1 is an uncontrollable field view environment, and interference of various factors such as stray light, water vapor, foreign matters and the like may exist, so that the image quality is low. The laser line is a target to be extracted from the image, but the above interference factors may cause problems such as unclear and intermittent laser lines. The image enhancement module 1 performs noise reduction and enhancement treatment on an original image of the surface of the strip steel through a specific image processing algorithm, highlights a detected object and weakens an interference object;
and the jitter judging and removing module 2 is used for judging the position and the accuracy of the side waves to be removed in time if the jitter is not removed in time because the jitter of the strip steel is caused by the production environment in the actual production process of the site. The jitter judging and removing module 2 is used for further judging the jitter phenomenon generated by the field production overload on the basis of the image enhancement module 1; if the jitter exists, the jitter is removed, and the image is continued to be processed in the next step.
The laser line extraction module 3 is used for obtaining the accurate position of the laser line in the image by adopting the image obtained by the processing of the previous steps and through algorithms such as gradient binarization, image segmentation or barycenter calculation, and the accuracy of laser line extraction determines the accuracy of final edge wave size and position calculation;
the laser line correction module 4, because the high-speed camera is arranged obliquely, the distances from the points at different positions on the strip steel to the imaging element are different, so the actual resolution of different pixel points on the acquired image is different. As a measurement system, the actual image resolution of each region must be known precisely so that the pixels can be converted into length units. In addition, actual imaging often has distortions such as the usual pincushion and barrel distortion due to the processing limitations of the lens. Distortion can also lead to inaccurate measurements. The image scale change caused by the above factors can obtain an accurate correction coefficient matrix through the earlier calibration step. The laser line correction module 4 performs inverse transformation on the laser lines in the image based on the determined matrix to acquire high-precision spatial profile data;
and the three-dimensional morphology reconstruction module 5 is used for calculating coordinate point positions of N (N is more than or equal to 2) laser line areas on the strip steel only in each image in online detection. Obviously, it is not possible to calculate the three-dimensional pattern of the entire strip with these N contours. An automatically updated data queue is therefore required, which records contour data extracted from the last several images. The three-dimensional morphology reconstruction module 5 performs vibration filtering on the spatial profile data, then performs merging treatment, draws the spatial profile data in a three-dimensional coordinate system to obtain a three-dimensional lattice diagram of the surface of the strip steel, and finally obtains three-dimensional graphic information of the surface of the whole strip steel through calculation and simulation;
the three-dimensional defect detection and identification module 6, the three-dimensional image information is not the final detection result, and the system needs to automatically judge the occurrence position of three-dimensional defects such as side waves. The three-dimensional defect detection and identification module 6 judges whether the strip steel edge wave defect exists or not from the three-dimensional graphic information, and determines defect position equalization information.
The image enhancement module 1 performs noise reduction and enhancement processing on an original image by using a gaussian filter algorithm and a histogram equalization image processing algorithm;
the Gaussian filter algorithm adopts a 5×5 Gaussian template, and the specific processing procedure is as follows:
1) Firstly calculating a Gaussian filter kernel of a Gaussian template according to the probability of standard two-dimensional normal distribution;
2) Coinciding each pixel point of the original image with the center point of the Gaussian filter kernel;
3) Multiplying and summing the original image and two 5x5 matrix corresponding elements overlapped by the Gaussian filter kernel, dividing the sum by a value obtained by normalizing the parameter 273, and replacing the value of the original image position;
the specific processing procedure of the histogram equalization image processing algorithm is as follows:
1) Scanning each pixel of the gray level of the original image in turn, and calculating a gray level histogram of the image;
2) Calculating a cumulative distribution function of the gray level histogram;
3) Obtaining a mapping relation between input and output according to the cumulative distribution function and the histogram equalization principle;
4) And obtaining a result according to the mapping relation and carrying out image transformation.
In field applications, the phenomenon of jitter in the production of strip steel can cause significant image problems. Therefore, the invention adopts the jitter judging and removing module 2, and the jitter judging and removing module 2 realizes the elimination of the jitter repeated lines existing in the image through image processing. The specific method comprises the following steps: in practical design, a fixed number of overlapping laser lines is set for the front and rear images. If the number of the laser lines which are overlapped by the front image and the rear image is increased or reduced (combined with the judgment of the background environment of the picture) in the detection process through image comparison, the strip steel is considered to have jitter in the motion process. And eliminating repeated laser line images, and avoiding error calculation of defect sizes and positions.
The laser line extraction module adopts a gray-scale gravity center laser line extraction algorithm to accurately extract the laser line.
The gray level gravity center laser line extraction algorithm is to process the gray level distribution characteristics in the cross section of each line of light stripe line by line, calculate and extract the gray level gravity center point of the light stripe area line by line in the direction of line coordinates, use the gray level gravity center point to represent the center point position of the light stripe of the cross section, and finally fit all the center points to form the center line of the light stripe;
the gray center of the object S in the gray image I (I, j) is (x) 0 ，y 0 ) The method comprises the following steps:
in the scheme, the specific operation steps are that filtering is firstly carried out on an image to remove noise, then threshold segmentation is carried out, a part larger than the threshold is reserved, namely an ROI (region of interest) area, and then a gray-scale gravity center method is used for extracting a laser center line in the area.
The gray-scale gravity center laser line extraction algorithm firstly adopts the maximum value to initially extract the center of the laser line, and then realizes the extraction precision of the laser line by weighting the gray-scale gravity center. Meanwhile, the width of the laser line is limited by combining with the field test, the line width pixel number W of the laser line is more than or equal to 8, namely, the width D of the laser line emitted by the laser device is required to meet D more than or equal to 8D on the assumption that the width direction of the image resolution required by detection is D (mm/pixel).
With continued reference to fig. 1, the invention also provides a system for processing and recognizing the three-dimensional detection image of the edge wave of the hot-rolled strip steel, which comprises a detection triggering unit 7, a laser structure light irradiator 8, a high-speed camera 9, an optical system view field adjusting unit 10, an image acquisition/transmission unit 11, an image storage/display unit 12 and a three-dimensional detection image processing and recognizing unit 13 of the edge wave of the hot-rolled strip steel.
The detection triggering unit 7 is used for capturing the signal of the strip steel on the roller way and triggering the starting of the laser structure light irradiator 8 and the high-speed camera 9.
A laser structured light irradiator 8 is installed above the strip steel and emits a laser line for illuminating the surface of the strip steel.
A high speed camera 9 is also mounted above the strip to capture images of the edge surface of the strip.
The optical system field adjusting unit 10 is used for adjusting the field of view of the high-speed camera 9 to achieve the adjustment of the angle θ between the field of view of the high-speed camera 9 and the laser line.
The image acquisition/transmission unit 11 is used for acquiring the edge surface image of the high-speed camera 9 and transmitting the edge surface image to the hot-rolled strip edge wave three-dimensional detection image processing and recognition unit 13. The image acquisition/transmission unit 11 is an image acquisition card.
The three-dimensional detection image processing and identifying unit 13 processes the edge surface image, calculates and identifies the wave distance and height of the upper edge wave of the hot rolled strip steel, and transmits the image to the image storage/display unit 12. The image processing and recognizing unit 13 is an image processing computer, and data communication is established between the image acquisition/transmission unit 11 and the image processing and recognizing unit 13.
The image storage/display unit 12 is used to enable storage of the edge surface image and alarm. The image storage/display unit 12 includes a data server and a terminal computer with which communication is established.
The detection triggering unit 7 comprises a photoelectric correlation switch and an encoder which are arranged on the roller way; the photoelectric opposite-shooting switch is arranged at a certain distance from the high-speed camera 9 along the roller way direction, and when strip steel passes through, the signal of the photoelectric opposite-shooting switch triggers the detection trigger unit 7 to start working, the laser structure light irradiator 8 is lightened, and the high-speed camera 9 starts photographing.
Considering the requirements of the strip production speed and image resolution on the exposure time of the high speed camera 9, this is a prerequisite for clear imaging. Assuming that the strip production speed is v (unit mm/s), the image imaging resolution single-pixel design accuracy is h (unit mm, strip movement direction) ×w (unit mm, strip width direction), the camera minimum exposure time is t=h/v.
The laser structure light irradiator 8 emits N (N is more than or equal to 2) laser lines, so that jitter interference in the production process can be effectively avoided. In the actual production process of the field, if only 1 laser line is adopted, if the shaking phenomenon exists in the production process, the strip steel at the same position can be photographed for multiple times, the image photographed by the single laser line lacks reference lines, the shaking lines cannot be removed during image processing, the strip steel image distortion is easy to be caused, and the problems of strip steel length calculation error increase, edge wave defect misjudgment and the like can be caused. N (N is more than or equal to 2) parallel laser lines are adopted for irradiation, the interval of the laser lines is d (unit mm), the imaging maximum view field of the camera along the strip steel moving direction is d (N-1) (unit mm), and the minimum frame rate of the camera isIn practical imaging design, because the strip steel jitter object is fully considered, the front image and the rear image need to be properly overlapped in the strip steel moving direction so as to facilitate the subsequent jitter filtering. Assuming that the effective field of view of the camera along the strip steel moving direction is designed to be d (M-1) (unit mm,2 is more than or equal to M is less than or equal to N-1), the overlapping area of the front image and the rear image is that the front image and the rear image have (N-M) contour lines which are repeated. At this time, the frame rate of the high-speed camera 3 is increased to +.>The detection requirement can be met.
The invention also provides a hot-rolled strip edge wave three-dimensional detection image processing and identifying method, which adopts the hot-rolled strip edge wave three-dimensional detection image processing and identifying system, the detection triggering unit 7 captures the strip passing signal and triggers the laser structure light irradiator 8 and the high-speed camera 9 to start working, the laser structure light irradiator 8 emits N parallel equidistant and vertical laser lines of the strip, the optical system view field adjusting unit 10 firstly adjusts the angle theta between the laser lines and the view field of the high-speed camera 9, the high-speed camera 9 shoots the edge surface image of the strip, the image collecting/transmitting unit 11 is used for collecting the edge surface image and transmitting the edge surface image to the hot-rolled strip edge wave three-dimensional detection image processing and identifying unit 13, the hot-rolled strip edge wave three-dimensional detection image processing and identifying unit 13 processes the edge surface image, calculates and identifies the wave distance and the height of the edge wave on the strip, and transmits the edge wave to the image storing/displaying unit 12, and the image storing/displaying unit 12 is used for realizing image storage and alarm.
Referring to fig. 2, the processing and recognizing unit 13 processes the edge surface image of the hot rolled strip as follows:
after the image acquired by the image acquisition/transmission unit 11 is transmitted to the hot-rolled strip edge wave three-dimensional detection image processing and recognition unit 13, the image enhancement module 1 firstly performs noise reduction and enhancement processing on the edge surface image. On the basis, the shake determination and removal module 2 performs shake determination on the edge surface image, and eliminates the shake phenomenon existing. Based on the above two processing flows, the laser line extraction module 3 extracts laser lines in the edge surface image. Then, the laser line correction module 4 eliminates distortion in the edge surface image and performs inverse transformation on the laser line to acquire high-precision spatial profile data. And the three-dimensional morphology reconstruction module 5 performs vibration filtering on the spatial profile data, then performs merging treatment, draws the spatial profile data in a three-dimensional coordinate system to obtain a three-dimensional lattice diagram of the surface of the strip steel, and obtains three-dimensional graphic information of the surface of the whole strip steel through calculation and simulation. Finally, the three-dimensional defect detection and identification module 6 judges whether the strip steel edge wave defect exists or not from the three-dimensional graphic information, and determines the defect position.
The processing and identifying unit, the system and the method for the three-dimensional detection image of the hot rolled strip steel edge wave can realize the three-dimensional accurate detection of the red hot strip steel edge wave, discover edge wave defects which possibly affect the subsequent production in time, reduce the next finishing process, greatly reduce the generation of quality objections, improve the production efficiency and have great economic and social benefits.
It will be appreciated by persons skilled in the art that the above embodiments are provided for illustration only and not for limitation of the invention, and that variations and modifications of the above described embodiments are intended to fall within the scope of the claims of the invention as long as they fall within the true spirit of the invention.
Claims (10)
1. The utility model provides a hot rolled strip limit unrestrained three-dimensional detection image processing and recognition element which characterized in that includes:
the image enhancement module is used for carrying out noise reduction and enhancement treatment on the original image on the surface of the strip steel;
the jitter judging and removing module is used for judging jitter phenomenon generated by field production overload;
the laser line extraction module is used for obtaining the accurate position of the laser line in the image;
the laser line correction module is used for carrying out inverse transformation on the laser lines in the image so as to obtain space contour data;
the three-dimensional morphology reconstruction module is used for carrying out vibration filtering on the spatial profile data, then merging the spatial profile data, drawing the spatial profile data in a three-dimensional coordinate system to obtain a three-dimensional lattice diagram of the surface of the strip steel, and finally obtaining three-dimensional graphic information of the surface of the whole strip steel through calculation and simulation;
and the three-dimensional defect detection and identification module is used for judging whether the strip steel edge wave defect exists or not from the three-dimensional graphic information and determining the defect position.
2. The three-dimensional detection image processing and recognition unit for the edge waves of the hot rolled strip according to claim 1, wherein: the image enhancement module performs noise reduction and enhancement processing on an original image by using a Gaussian filtering algorithm and a histogram equalization image processing algorithm;
the Gaussian filtering algorithm adopts a 5×5 Gaussian template, and the specific processing procedure is as follows:
1) Firstly calculating a Gaussian filter kernel of the Gaussian template according to the probability of standard two-dimensional normal distribution;
2) Coinciding each pixel point of the original image with a center point of the Gaussian filter kernel;
3) Multiplying and summing the original image and two 5x5 matrix corresponding elements overlapped by the Gaussian filter kernel, dividing the sum by a value obtained by normalizing parameter 273, and replacing the value of the original image position;
the specific processing procedure of the histogram equalization image processing algorithm is as follows:
1) Scanning each pixel of the original image gray level in turn, and calculating a gray level histogram of the image;
2) Calculating a cumulative distribution function of the gray level histogram;
3) Obtaining a mapping relation between input and output according to the cumulative distribution function and the histogram equalization principle;
4) And obtaining a result according to the mapping relation and carrying out image transformation.
3. The three-dimensional detection image processing and recognition unit for the edge waves of the hot rolled strip according to claim 1, wherein: the jitter judging and removing module is used for removing jitter repeated lines existing in the image through image processing.
4. The three-dimensional detection image processing and recognition unit for the edge waves of the hot rolled strip according to claim 1, wherein: the laser line extraction module adopts a gray-scale gravity center laser line extraction algorithm to accurately extract the laser line;
the gray-scale gravity center laser line extraction algorithm is to process the gray-scale distribution characteristics in the cross section of each row of light stripes row by row, calculate and extract gray-scale gravity center points of the light stripe areas row by row in the direction of row coordinates, use the gray-scale gravity center points to represent the center point positions of the light stripes of the cross section, and finally fit all the center points to form the center line of the light stripes.
5. The three-dimensional detection image processing and recognition unit for the edge waves of the hot rolled strip according to claim 4, wherein: the gray-scale gravity center laser line extraction algorithm firstly adopts the maximum value to initially extract the center of the laser line, and then realizes the extraction precision of the laser line through weighting the gray-scale gravity center.
6. The three-dimensional detection image processing and recognition unit for the edge waves of the hot rolled strip according to claim 4, wherein: the line width pixel number W of the laser line is more than or equal to 8.
7. A three-dimensional detection image processing and identifying system for hot rolled strip steel edge wave is characterized in that: comprises a detection triggering unit, a laser structure light irradiator, a high-speed camera, an optical system view field adjusting unit, an image acquisition/transmission unit, an image storage/display unit and a hot-rolled strip edge wave three-dimensional detection image processing and identifying unit according to one of claims 1 to 6;
the detection triggering unit is used for capturing the signal of the strip steel on the roller way and triggering the starting of the laser structure light irradiator and the high-speed camera;
the laser structure light irradiator is arranged above the strip steel and emits laser rays to illuminate the surface of the strip steel;
the high-speed camera is also arranged above the strip steel and is used for shooting an edge surface image of the strip steel;
the optical system view field adjusting unit is used for adjusting the view field of the high-speed camera;
the image acquisition/transmission unit is used for acquiring the edge surface image of the high-speed camera and transmitting the edge surface image to the hot-rolled strip steel edge wave three-dimensional detection image processing and identifying unit;
the hot rolled strip steel edge wave three-dimensional detection image processing and identifying unit processes the edge surface image, calculates and identifies the wave distance and the height of the strip steel upper edge wave, and transmits the wave distance and the height to the image storage/display unit;
the image storage/display unit is used for realizing storage and alarm of the edge surface image.
8. The three-dimensional inspection image processing and recognition system for edge waves of hot rolled strip according to claim 7, wherein: the detection triggering unit comprises a photoelectric correlation switch and an encoder which are arranged on the roller way;
the number N of the laser lines is more than or equal to 2.
The minimum exposure time t=h/v of the high-speed camera, wherein h is v which is the production speed of the belt steel;
the image acquisition/transmission unit is an image acquisition card;
the hot-rolled strip steel edge wave three-dimensional detection image processing and identifying unit is an image processing computer;
the image storage/display unit includes a data server and a terminal computer in communication therewith.
9. The three-dimensional inspection image processing and recognition system for edge waves of hot rolled strip according to claim 7, wherein: the laser structure light irradiator, the high-speed camera and the optical system view field adjusting unit are all arranged in the protective box.
10. A three-dimensional detection image processing and identifying method for hot rolled strip steel edge waves is characterized in that: the three-dimensional detection image processing and identifying system for the edge wave of the hot rolled strip steel is adopted, the detection triggering unit captures a signal of the strip steel passing through and triggers the laser structure light irradiator and the high-speed camera to start, the laser structure light irradiator emits N laser lines which are parallel and equidistant and are perpendicular to the strip steel, the optical system view field adjusting unit firstly adjusts the angle theta between the laser lines and the view field of the high-speed camera, the high-speed camera shoots an edge surface image of the strip steel, the image acquisition/transmission unit is used for acquiring the edge surface image and transmitting the edge surface image to the three-dimensional detection image processing and identifying unit for the edge surface image of the hot rolled strip steel, the three-dimensional detection image processing and identifying unit for the edge surface image processes, calculates and identifies the wave distance and the height of the edge wave on the strip steel and transmits the image to the image storage/display unit, and the image storage/display unit is used for realizing the storage and alarm of the image;
the hot-rolled strip steel edge wave three-dimensional detection image processing and identifying unit processes the edge surface image as follows:
the image enhancement module performs noise reduction and enhancement processing on the edge surface image, the jitter judgment and removal module performs jitter judgment on the edge surface image and removes the existing jitter phenomenon, the laser line extraction module extracts laser lines in the edge surface image, the laser line correction module eliminates distortion in the edge surface image and performs inverse transformation on the laser lines so as to obtain space contour data, the three-dimensional shape reconstruction module performs vibration filtering on the space contour data and then merges the space contour data, the space contour data is drawn in a three-dimensional coordinate system to obtain a three-dimensional lattice diagram of the strip steel surface, finally three-dimensional graph information of the whole strip steel surface is obtained through calculation and simulation, the three-dimensional defect detection recognition module judges whether the strip steel edge wave defect exists or not from the three-dimensional graph information and determines the defect position.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210804607.XA CN117422664A (en) | 2022-07-08 | 2022-07-08 | Processing and identifying unit, system and method for three-dimensional detection image of hot rolled strip steel edge wave |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210804607.XA CN117422664A (en) | 2022-07-08 | 2022-07-08 | Processing and identifying unit, system and method for three-dimensional detection image of hot rolled strip steel edge wave |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117422664A true CN117422664A (en) | 2024-01-19 |
Family
ID=89531241
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210804607.XA Pending CN117422664A (en) | 2022-07-08 | 2022-07-08 | Processing and identifying unit, system and method for three-dimensional detection image of hot rolled strip steel edge wave |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117422664A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN119063650A (en) * | 2024-11-06 | 2024-12-03 | 中铁四局集团有限公司 | Method and system for detecting and positioning bladder expansion based on machine vision |
-
2022
- 2022-07-08 CN CN202210804607.XA patent/CN117422664A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN119063650A (en) * | 2024-11-06 | 2024-12-03 | 中铁四局集团有限公司 | Method and system for detecting and positioning bladder expansion based on machine vision |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109190628A (en) | A kind of plate camber detection method based on machine vision | |
| CN109559324B (en) | A method of target contour detection in line image | |
| CN107369159B (en) | Threshold segmentation method based on multi-factor two-dimensional gray level histogram | |
| CN108088845A (en) | A kind of image-forming correction method and device retained based on Weak Information | |
| CN109978874B (en) | Visual detection device and identification method for surface defects of steel rail | |
| CN104048969A (en) | Tunnel defect recognition method | |
| CN111421001B (en) | Online accurate centering system of high-speed wire rod mill and centering method thereof | |
| CN105203552A (en) | 360-degree tread image detecting system and method | |
| CN111521129A (en) | Machine vision-based slab warpage detection device and method | |
| WO2022233096A1 (en) | Method and apparatus for optimal shearing control of steel billet based on depth vision | |
| CN111354047B (en) | Computer vision-based camera module positioning method and system | |
| CN104458764B (en) | Curved uneven surface defect identification method based on large-field-depth stripped image projection | |
| CN114170165A (en) | Chip surface defect detection method and device | |
| CN115018735B (en) | Crack width identification method and system based on Hough transformation correction two-dimensional code image | |
| WO2023134251A1 (en) | Clustering-based light bar extraction method and apparatus | |
| CN112949482A (en) | Non-contact type rail sleeper relative displacement real-time measurement method based on deep learning and visual positioning | |
| CN115953550A (en) | Point cloud outlier removal system and method for line structured light scanning | |
| CN108510544B (en) | A method of light strip location based on feature clustering | |
| CN109671084B (en) | A method for measuring workpiece shape | |
| CN111524134B (en) | Detection method and detection device for regularity of cellular products on production line | |
| CN117422664A (en) | Processing and identifying unit, system and method for three-dimensional detection image of hot rolled strip steel edge wave | |
| CN113819841B (en) | Plate shape detection device based on machine vision and detection method thereof | |
| JP2022181167A (en) | Shape determination method of steel plate, shape measurement method, shape control method, manufacturing method, generation method of shape determination model and shape determination device | |
| CN113409334A (en) | Centroid-based structured light angle point detection method | |
| KR102161315B1 (en) | Automatic identification monitoring system of container |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |