CN102441581A - Machine vision-based device and method for online detection of structural steel section size - Google Patents
Machine vision-based device and method for online detection of structural steel section size Download PDFInfo
- Publication number
- CN102441581A CN102441581A CN2010102974147A CN201010297414A CN102441581A CN 102441581 A CN102441581 A CN 102441581A CN 2010102974147 A CN2010102974147 A CN 2010102974147A CN 201010297414 A CN201010297414 A CN 201010297414A CN 102441581 A CN102441581 A CN 102441581A
- Authority
- CN
- China
- Prior art keywords
- image
- sectional shape
- shaped steel
- camera
- structural steel
- 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
Images
Abstract
The invention discloses a machine vision-based technology of a structural steel section, belonging to the field of online detection of the structural steel section size in the metallurgy industry. The invention is characterized in that hardware of a device comprises a camera, an induction switch, an image acquisition card, a personal computer and the like, software of the device comprises the steps of: image smoothing, edge extracting, linear circle detecting, coordinate transforming, geometry correcting and the like, an actual size of structural steel is figured out through an image of the structural steel section, obtained by the camera, and a standard object in the image, and finally whether the structural steel accords with the production requirement is judged. The method has the advantages of good instantaneity, rapidness and accuracy, low cost and strong practicability.
Description
Technical field
The present invention relates to the field of the online detection of metallurgy industry sectional shape size, specifically belong to device and method based on the online detection of sectional shape size of machine vision (collecting image of computer and processing).
Background technology
In hot-rolled steel section is produced; Because high temperature causes the difficulty of the online detection of product cross dimensions, in cold-rolled forming section is produced, because section configuration is complicated; Test point many and be the difficult point of cold-rolled forming section on-line monitoring always; Manual inspection wastes time and energy, and conventional online detecting instrument test point is few, can not detect the physical dimension of the whole section of being produced of product.Though existing a kind of laser on-line detector device; Utilize the principle of laser ranging, product surface is scanned, size that can the whole section of testing product with laser beam; But no matter this device is IMAQ sensor or lasing light emitter; Complex structure is with high costs, and its software also depends on professional module, has therefore restricted popularizing of laser measuring technology.
With popularizing of computer technology; Machine vision is used widely, but does not also use at the sectional shape detection range, according to retrieving invention disclosed patent; Application number is 200910073924.3 " based on the steel blank regrinding grinding wheel diameter checkout gears of machine vision "; Though also be Machine Vision Detection, its device is also similar, and the present invention has following different with correlation technique: 1, detected object is different; The detected object abrasive machine of correlation technique, and detected object of the present invention is the shaped steel of production line.2, the test point of correlation technique is this index of displacement of emery wheel edge abrasion, and the content that the present invention detects relates to the overall dimension of sectional shape, comprises the height of product, width, and thickness, the radius of each curved radius etc., it is more complicated to detect index.3, accuracy of detection requires difference, and grinding wheel diameter grinds little and scraps, and normally used precision unit is a millimeter; Its software processes process is simple relatively, and linear regression and the least square method only used are obtained the determinand vary in diameter.And software section of the present invention has adopted multiple processing mode to guarantee precision; Confirm correction value etc. like multiple image filtering processing, twice linear regression, experiment; Its accuracy of detection can reach 0.1 millimeter in theory, so level of technical sophistication of the present invention is higher relatively.
Summary of the invention
The objective of the invention is to utilize popular universal PC and camera (or video camera) and develop sectional shape size line Measurement Technique based on machine vision.The present invention utilizes shaped steel continuous in the steel rolling production process after flying to saw the scale sawing, and the ad-hoc location in type steel production line is provided with inductive switch; The mode of continuously taking pictures with camera obtain each product section and benchmark scale image and be transported to computer, through software processing, sectional shape figure in the image and benchmark lines are extracted simultaneously; And change into polar plot, calculate the similar physical dimension of each point of shaped steel with geometrical relationship, and the shaped steel image and the benchmark lines that obtain are carried out true ratio adjustment; The image that obtains is in real time compared; Reach monitoring, forecast model products quality, the purpose of reduction defective products rate.
Form by hardware and software two parts based on the sectional shape size on-line measuring device of machine vision and the characteristic of method; Hardware characteristics is: by camera (or video camera) (1); Inductive switch (4); Personal computer (2) and stube cable etc. constitute hardware components, constitute its software section by image processing program.
Sectional shape size on-line measuring device and method hardware components characteristic based on machine vision are: inductive switch (4); Link to each other with the photographing switch of (or shooting) machine (1) of taking a picture; Camera (1) is connected with a complete personal computer (2); The same position that arrives in shot region and shaped steel termination has the benchmark graticule (6) of a preseting length, and this benchmark graticule and examined product can get into the image of picked-up simultaneously; For increasing the contrast of image chroma, be provided with shade (5) at the shaped steel surveyed area.
Image processing program (3) (in computer (2) inside) is used for the image that receives is carried out a series of processing, detection, judgement, and finally draws whether standard compliant judged result of sectional shape.
Sectional shape size on-line measuring device and method software section based on machine vision are characterised in that the employing following steps:
Step 1: set the canonical parameter and the Measurement Allowance of sectional shape in advance, be stored in the computer with the mode of vector data.Parameter comprises length and width, the arc radius of sectional shape.
Step 2: certain that pass through at shaped steel is a bit taken pictures once by inductive switch control camera, obtains to comprise the examined product cross-section image of datum line.
Step 3: the camera image that obtains of taking pictures imports in the calculator memory through image pick-up card.
Step 4: program is carried out the sharpening enhancing contrast ratio to the positive image of the sectional shape in the internal memory.
Step 5: above-mentioned result is carried out picture smooth treatment
Step 6: then above result is carried out Threshold Segmentation.Purpose is the part that from image, extracts sectional shape.
Step 7: the bianry image that above processing obtains carries out morphological operation: fill little cavity, separate little isolated point, ON operation.
Step 8: the result that above processing is obtained is converted into gray level image, carries out smothing filtering, and purpose is to make the edge regular.
Step 9: above process result is carried out edge extracting, obtain the image border of sectional shape.
Step 10: to the bianry image detection of straight lines section at edge.
Step 11: line segment is pressed slope sort out, then the line segment end points is done linear regression, draw the linear equation on border
Step 12: the bianry image to the edge is justified detection, obtain shaped steel external boundary circular arc portion the center of circle and radius.Coordinate system is the location of pixels of image
Step 13: the confirming of image distortion correcting proportion coefficient.
Step 14:, coordinate transform is carried out on the border of shaped steel according to the Pixel Dimensions and the actual size of standard component.Obtain the shaped steel polar plot model of actual ratio.
Step 15: calculate the wide of shaped steel, height, radius of corner critical size parameter with geometry.
Step 16: the confirming of correction value: systematic error is confirmed in the comparison through a large amount of measured values and actual value.Systematic error is set at correction value.
Step 17: the shaped steel critical size parameter of using step 15 to calculate, add correction value, with the standard form comparison of prior input, confirm whether the shaped steel dimensional parameters meets standard.
The specific embodiment
Be the specific embodiment of example explanation with the clod wash square tube below based on the sectional shape detection technique of machine vision.At first in the process that shaped steel is produced, the steel after the sawing can stop the of short duration time in this system's detection place.The operator sends the instruction of taking pictures at this moment through computer 3; Because the dark veil 6 of background; Sectional shape can demonstrate brighter color in photo, and background and shaped steel inside can demonstrate dark color, presents the full-length mark on also the having powerful connections of Gao Liang simultaneously.Camera 1 on the production line imports the photo of this sectional shape that photographed in real time into computer 3 through image pick-up card 2, temporarily stores with the form of colored bitmap.Software 4 is handled like following steps then:
With reference to figure 2-S1: the parameter of importing shaped steel in advance has standard length and width, arc radius, the Measurement Allowance of section.
With reference to figure 2-S2, S3: inductive switch control camera is taken pictures to sectional shape during shaped steel process surveyed area.Computer obtains the photo bitmap of section.
With reference to figure 2-S4: software to image sharpening to eliminate the interference of part shade.The algorithm of sharpening is chosen according to actual working environment.Purpose is to increase contrast to be convenient to image and to cut apart.
With reference to figure 2-S5: because Image Sharpening Algorithm can produce noise in image.Therefore continue to last step process result carry out smothing filtering (the specific algorithm view is fixed as situation, such as this example use be exactly gaussian filtering with the elimination noise, also can use simple mean filter).
With reference to figure 2-S6: the method with Threshold Segmentation extracts the sectional shape part.Threshold Segmentation has following two kinds of methods, looks concrete image situation and decides.
Directly be converted into 256 grades of gray scales to image with reference to figure S-61, threshold value of gray scale regulation is thought sectional shape greater than the pixel of this value, less than think background.This algorithm is used for the higher image of contrast.
With reference to figure S-62 because the image color characteristic of sectional shape is that three kinds of color components of RGB differ smaller (being mostly white or grey), the rgb value of each pixel is calculated three kinds of color components of RGB and its mean value difference and.Then think sectional shape as if this with less than certain threshold value, then think background greater than threshold value.
With reference to figure 2-S7: because Threshold Segmentation is not 100% accurate, has partial pixel to have few part in the actual sectional shape and can be considered to background and therefore produce the cavity, perhaps the point of background is thought sectional shape and is produced isolated point.Therefore need carry out isolated point eliminates and fills.Filling algorithm is that area is judged.Be filled to the isolated point of less area the color on border.Carry out the morphology ON operation at last, purpose is to make the edge further regular.The result who obtains is a black and white binary image.
With reference to figure 2-S8: because still there are a lot of relatively large jagged edges in the bianry image that top step obtains.Therefore need make these edges further level and smooth.Here adopt and convert gray level image smothing filtering again into.This example adopts the 4*4 mean filter.
With reference to figure 2-S9: the edge extracting that carries out sectional shape.There are two kinds of algorithms to adopt: morphologic remove algorithm and based on the edge extracting of canny operator.Use the former more suitable for enough regular edge.And be applicable to the latter for the edge of irregularity.The latter that this example adopts.
With reference to figure 2-S10: use the straightway in the hough change detection image.Specific algorithm thought is: the edge point set of the object in the digital picture all is a discrete data, makes that the image diagonal distance is d, can realize the Hough conversion by following step according to above-mentioned principle:
(1) with θ-ρ parameter space be quantified as m*n (wherein m is the umber that waits of θ, and n is the umber that waits of ρ, 0<=θ<180 degree ,-d<=ρ<=-d) individual unit, corresponding to setting up accumulator matrix T (m*n) in each unit;
(2) all accumulator initial values are put 0;
(3) in 0<=θ<180 are interval, get θ successively to the every bit in the detected point set, calculate corresponding ρ value, and add 1 at corresponding accumulator element in view of the above by quantizing step-length, and collection T (i, j)=T (i, j)+1;
(4) scanning accumulator matrix, T (i, j) value maximum (θ, ρ) the corresponding straight line of value be ask;
(5) when many straight lines are arranged in the image, with the accumulator matrix extracted straight line (θ, ρ) accumulator value of regional area puts 0 near the value, changes (4), up to finding out all straight lines.
With reference to figure 2-S11: because edge quality, often detected is many line segments on the edge line.These line segments belong in several the straight lines.Therefore press slope to the end points of these line segments again and classify, slope is close just thinks line segment on same the straight line, and the line segment end points to same straight line carries out linear regression then, confirms the linear equation on shaped steel standard component border.This moment, coordinate system was the location of pixels of image
With reference to figure 2-S12: to the bianry image at edge justify detection (using improved hough conversion) obtain shaped steel external boundary circular arc portion the center of circle and radius.The specific algorithm principle is following:
The general equation of known circle is: (x-a) 2+ (y-b) 2=r2
Wherein: (a b) is the center of circle, and r is a radius of a circle.Be transformed into the a-b-r parameter space to the circle on the X-Y plane, then the circle of any is corresponding to a three-dimensional conical surface in the parameter space excessively arbitrarily in the image space, and the point in the image space on the same circle must meet at a bit corresponding to all the three-dimensional conical surfaces in the parameter space.Through detecting the parameter that this point can obtain circle, corresponding circle can be tried to achieve like this.
With reference to figure 2-S13: because camera can not be placed on the passage of shaped steel, the angle of camera axis and sectional shape is not strict 90 to spend when therefore taking pictures.Specific algorithm does, surveys out the angle on camera axis and sectional shape plane, and using length that trigonometric function calculates actual projection and angle is the proportionality coefficient of the length of projection under 90 normal conditions spent.
With reference to figure 2-S14: because this moment, coordinate system was a pixel coordinate, the parasang that therefore calculates also is a number of pixels, need convert pixel coordinate into actual coordinate now.Principle is the ratio conversion according to pixel distance and actual range.Specific practice is the straight line pixel distance of standard component actual range divided by the edge spacing of detected standard component, obtains proportionality coefficient r.Then the known pixel coordinate coordinate system of the coefficient r actual size of dwindling in proportion.
With reference to figure 2-S15: calculate directly and the distance between straight line with the range formula between straight line.Wherein the rectilineal interval of the straight line of X coordinate maximum and X coordinate minimum is from the height that is exactly shaped steel, and the distance between the straight line that the Y coordinate is maximum and the Y coordinate is minimum is exactly the width of shaped steel.
With reference to figure 2-S16: because in the process that image is handled, filtering, morphologic various operations can make the edge micro-deformation.Therefore need relatively more definite systematic error through a large amount of measured values and actual value.Confirm correction value according to systematic error.(carrying out this step when only moving debugging for the first time) in system
With reference to figure 2-S17:: the shaped steel critical size parameter (wide, height, radius of corner) of using step 2-S15 to calculate; Add correction value; Compare one by one with the standard value of prior input, if with the difference of standard value all in Measurement Allowance, assert that then shaped steel meets standard.Otherwise system alarm is also pointed out a certain standard that do not meet.
Cost of the present invention is low, and the detection size is comprehensive, detection efficiency is high, and precision can satisfy production requirement, is beneficial to and applies.
Description of drawings
Fig. 1: based on the sectional shape checkout gear hardware structure diagram of machine vision
1 camera
2 home computers
3 image processing softwares
4 inductive switches
5 background veils
6 size datum graticules
7 production line shaped steel rollgangs
Fig. 2: based on the sectional shape checkout gear software flow pattern of machine vision.
Claims (3)
1. form by hardware and software two parts based on the sectional shape size on-line measuring device of machine vision and the characteristic of method; Hardware characteristics is: by camera (or video camera) (1); Inductive switch (4); Personal computer (2) and stube cable etc. constitute hardware components, constitute its software section by the Computer Image Processing program.
2. sectional shape size on-line measuring device and its hardware characteristics of method based on machine vision according to claim 1 is: inductive switch (4); Link to each other with the photographing switch of (or shooting) machine (1) of taking a picture; Camera (1) is connected with a complete personal computer (2); The same position that arrives in shot region and shaped steel termination has the benchmark graticule (6) of a preseting length, is provided with shade (5) at the shaped steel surveyed area.
3. sectional shape size on-line measuring device and the method software section based on machine vision according to claim 1 is characterised in that the employing following steps:
Step 1: set the canonical parameter and the Measurement Allowance of sectional shape in advance, be stored in the computer with the mode of vector data, parameter comprises length and width, the arc radius of sectional shape;
Step 2: certain that pass through at shaped steel is a bit taken pictures once by inductive switch control camera, obtains to comprise the examined product image of datum line;
Step 3: the camera image that obtains of taking pictures imports in the calculator memory through image pick-up card;
Step 4: program is carried out sharpening to the positive image of the sectional shape in the internal memory;
Step 5: above-mentioned result is carried out picture smooth treatment;
Step 6: then above result is carried out the part that Threshold Segmentation extracts sectional shape;
Step 7: the bianry image that above processing obtains carries out morphological operation: fill little cavity, separate little isolated point, ON operation;
Step 8: the result that above processing is obtained is converted into gray level image, carries out smothing filtering and makes the edge regular;
Step 9: above process result is carried out edge extracting, obtain the image border of sectional shape;
Step 10: to the bianry image detection of straight lines section at edge;
Step 11: line segment is pressed slope sort out, then the line segment end points is done linear regression, draw the linear equation on border;
Step 12: to the bianry image at edge justify detect obtain shaped steel external boundary circular arc portion the center of circle and radius, coordinate system is the location of pixels of image;
Step 13: confirm image distortion correcting proportion coefficient;
Step 14:, coordinate transform is carried out on the border of shaped steel according to the Pixel Dimensions and the actual size of standard component.Obtain the shaped steel polar plot model of actual ratio;
Step 15: the critical size parameter (wide, height, radius of corner) of calculating shaped steel with geometry;
Step 16: the confirming of correction value: systematic error is confirmed in the comparison through a large amount of measured values and actual value.Systematic error is set at correction value;
Step 17: the shaped steel critical size parameter (wide, height, radius of corner) of using step 15 to calculate, add correction value, with the standard form comparison of prior input, confirm whether the shaped steel dimensional parameters meets standard.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2010102974147A CN102441581A (en) | 2010-09-30 | 2010-09-30 | Machine vision-based device and method for online detection of structural steel section size |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2010102974147A CN102441581A (en) | 2010-09-30 | 2010-09-30 | Machine vision-based device and method for online detection of structural steel section size |
Publications (1)
Publication Number | Publication Date |
---|---|
CN102441581A true CN102441581A (en) | 2012-05-09 |
Family
ID=46004802
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN2010102974147A Pending CN102441581A (en) | 2010-09-30 | 2010-09-30 | Machine vision-based device and method for online detection of structural steel section size |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN102441581A (en) |
Cited By (17)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102840839A (en) * | 2012-09-18 | 2012-12-26 | 重庆大学 | Accurate grinding wheel section measurement and error compensation method |
CN103292701A (en) * | 2013-06-24 | 2013-09-11 | 哈尔滨工业大学 | Machine-vision-based online dimensional measurement method of precise instrument |
CN105327950A (en) * | 2015-11-09 | 2016-02-17 | 北京科技大学 | Hot rolling intermediate billet camber detecting device and detecting method considering thickness change |
CN106706662A (en) * | 2016-11-17 | 2017-05-24 | 彩虹(合肥)液晶玻璃有限公司 | TFT-LCD glass substrate semi-finished product section analysis device |
CN107490342A (en) * | 2017-06-30 | 2017-12-19 | 广东工业大学 | A kind of cell phone appearance detection method based on single binocular vision |
CN107748897A (en) * | 2017-10-30 | 2018-03-02 | 南京工业大学 | Large scale crooked part profile tolerance quality determining method based on pattern-recognition |
CN110639965A (en) * | 2019-09-25 | 2020-01-03 | 中冶东方工程技术有限公司 | Weight measuring device, conveying device and method for steel rolling production |
CN110675393A (en) * | 2019-09-29 | 2020-01-10 | 上海呈彧智能科技有限公司 | Blank specification detection method based on machine vision |
CN111256596A (en) * | 2020-02-21 | 2020-06-09 | 北京容联易通信息技术有限公司 | Size measuring method and device based on CV technology, computer equipment and medium |
CN111524099A (en) * | 2020-04-09 | 2020-08-11 | 武汉钢铁有限公司 | Method for evaluating geometric parameters of cross section of sample |
CN112165996A (en) * | 2018-05-22 | 2021-01-01 | 东芝三菱电机产业系统株式会社 | Image analysis device for industrial plant and industrial plant monitoring control system |
CN113375555A (en) * | 2018-07-02 | 2021-09-10 | 广西电网有限责任公司北海供电局 | Power line clamp measuring method and system based on mobile phone image |
CN113490570A (en) * | 2019-02-27 | 2021-10-08 | 株式会社天田集团 | Machining estimation device, machining estimation method, and machining estimation program |
CN113533332A (en) * | 2021-08-10 | 2021-10-22 | 江西离子型稀土工程技术研究有限公司 | Praseodymium-neodymium alloy qualitative quality inspection method based on section image characteristics |
CN113607089A (en) * | 2021-06-23 | 2021-11-05 | 新乡市中智重工机械设备有限公司 | Section steel bending method based on vision measurement |
CN114777686A (en) * | 2017-10-06 | 2022-07-22 | 先进扫描仪公司 | Generating one or more luminance edges to form a three-dimensional model of an object |
CN115311629A (en) * | 2022-10-12 | 2022-11-08 | 南通创为机械科技有限公司 | Abnormal bending precision monitoring system of bending machine |
-
2010
- 2010-09-30 CN CN2010102974147A patent/CN102441581A/en active Pending
Cited By (24)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102840839A (en) * | 2012-09-18 | 2012-12-26 | 重庆大学 | Accurate grinding wheel section measurement and error compensation method |
CN103292701A (en) * | 2013-06-24 | 2013-09-11 | 哈尔滨工业大学 | Machine-vision-based online dimensional measurement method of precise instrument |
CN103292701B (en) * | 2013-06-24 | 2015-09-16 | 哈尔滨工业大学 | The online dimension measurement method of accurate device based on machine vision |
CN105327950A (en) * | 2015-11-09 | 2016-02-17 | 北京科技大学 | Hot rolling intermediate billet camber detecting device and detecting method considering thickness change |
CN106706662A (en) * | 2016-11-17 | 2017-05-24 | 彩虹(合肥)液晶玻璃有限公司 | TFT-LCD glass substrate semi-finished product section analysis device |
CN107490342A (en) * | 2017-06-30 | 2017-12-19 | 广东工业大学 | A kind of cell phone appearance detection method based on single binocular vision |
CN114777686A (en) * | 2017-10-06 | 2022-07-22 | 先进扫描仪公司 | Generating one or more luminance edges to form a three-dimensional model of an object |
CN107748897A (en) * | 2017-10-30 | 2018-03-02 | 南京工业大学 | Large scale crooked part profile tolerance quality determining method based on pattern-recognition |
CN112165996A (en) * | 2018-05-22 | 2021-01-01 | 东芝三菱电机产业系统株式会社 | Image analysis device for industrial plant and industrial plant monitoring control system |
CN112165996B (en) * | 2018-05-22 | 2022-07-29 | 东芝三菱电机产业系统株式会社 | Industrial complete equipment monitoring control system |
CN113375555A (en) * | 2018-07-02 | 2021-09-10 | 广西电网有限责任公司北海供电局 | Power line clamp measuring method and system based on mobile phone image |
CN113490570B (en) * | 2019-02-27 | 2024-03-19 | 株式会社天田集团 | Machining estimation device, machining estimation method, and machining estimation program |
US11914352B2 (en) | 2019-02-27 | 2024-02-27 | Amada Co., Ltd. | Machining estimation apparatus, machining estimation method, and machining estimation program |
CN113490570A (en) * | 2019-02-27 | 2021-10-08 | 株式会社天田集团 | Machining estimation device, machining estimation method, and machining estimation program |
CN110639965A (en) * | 2019-09-25 | 2020-01-03 | 中冶东方工程技术有限公司 | Weight measuring device, conveying device and method for steel rolling production |
CN110639965B (en) * | 2019-09-25 | 2020-11-10 | 中冶东方工程技术有限公司 | Weight measuring device, conveying device and method for steel rolling production |
CN110675393A (en) * | 2019-09-29 | 2020-01-10 | 上海呈彧智能科技有限公司 | Blank specification detection method based on machine vision |
CN110675393B (en) * | 2019-09-29 | 2024-03-26 | 上海呈彧智能科技有限公司 | Blank specification detection method based on machine vision |
CN111256596A (en) * | 2020-02-21 | 2020-06-09 | 北京容联易通信息技术有限公司 | Size measuring method and device based on CV technology, computer equipment and medium |
CN111524099A (en) * | 2020-04-09 | 2020-08-11 | 武汉钢铁有限公司 | Method for evaluating geometric parameters of cross section of sample |
CN113607089A (en) * | 2021-06-23 | 2021-11-05 | 新乡市中智重工机械设备有限公司 | Section steel bending method based on vision measurement |
CN113607089B (en) * | 2021-06-23 | 2023-11-10 | 新乡市中智重工机械设备有限公司 | Profile steel bending method based on visual measurement |
CN113533332A (en) * | 2021-08-10 | 2021-10-22 | 江西离子型稀土工程技术研究有限公司 | Praseodymium-neodymium alloy qualitative quality inspection method based on section image characteristics |
CN115311629A (en) * | 2022-10-12 | 2022-11-08 | 南通创为机械科技有限公司 | Abnormal bending precision monitoring system of bending machine |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN102441581A (en) | Machine vision-based device and method for online detection of structural steel section size | |
CN109141232B (en) | Online detection method for disc castings based on machine vision | |
CN110286124B (en) | Machine vision-based refractory brick measuring system | |
CN106969706A (en) | Workpiece sensing and three-dimension measuring system and detection method based on binocular stereo vision | |
CN102496161B (en) | Method for extracting contour of image of printed circuit board (PCB) | |
CN104458764B (en) | Curved uneven surface defect identification method based on large-field-depth stripped image projection | |
CN106767564A (en) | A kind of detection method for being applied to phone housing surface roughness | |
CN110717909A (en) | Metal surface scratch detection method and device | |
CN111539927A (en) | Detection process and algorithm of automobile plastic assembly fastening buckle lack-assembly detection device | |
CN104515479A (en) | Medium plate surface plate shape measuring system and method | |
CN112489042A (en) | Metal product printing defect and surface damage detection method based on super-resolution reconstruction | |
CN106989672A (en) | A kind of workpiece measuring based on machine vision | |
CN103337067B (en) | The visible detection method of single needle scan-type screw measurement instrument probe X-axis rotating deviation | |
CN110956624A (en) | Image definition evaluation method for three-dimensional object | |
Ali et al. | Application of machine vision in improving safety and reliability for gear profile measurement | |
CN113252103A (en) | Method for calculating volume and mass of material pile based on MATLAB image recognition technology | |
CN111539951B (en) | Visual detection method for outline size of ceramic grinding wheel head | |
CN101696876B (en) | Visual detection method for VCM magnetic steel | |
CN113607058B (en) | Straight blade size detection method and system based on machine vision | |
CN114549403A (en) | Multi-monomer intelligent precise geometric circle center detection method for side section of mechanical part | |
Ali et al. | Vision based measurement system for gear profile | |
CN108401142B (en) | Workpiece counting device and method on assembly line | |
CN113628155A (en) | Green ball particle size detection method and system of disc pelletizer | |
CN111210419A (en) | Micro magnetic tile surface defect detection method based on human visual characteristics | |
CN115582840B (en) | Method and system for calculating sorting and grabbing pose of borderless steel plate workpiece and sorting method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
DD01 | Delivery of document by public notice |
Addressee: Deng Yue Document name: Notification that Application Deemed to be Withdrawn |
|
C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20120509 |