CN104729413B - One kind is puted the palms together before one product width online automatic detection and adjusting means and method - Google Patents

One kind is puted the palms together before one product width online automatic detection and adjusting means and method Download PDF

Info

Publication number
CN104729413B
CN104729413B CN201510125641.4A CN201510125641A CN104729413B CN 104729413 B CN104729413 B CN 104729413B CN 201510125641 A CN201510125641 A CN 201510125641A CN 104729413 B CN104729413 B CN 104729413B
Authority
CN
China
Prior art keywords
width
product
mrow
together before
palms together
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.)
Active
Application number
CN201510125641.4A
Other languages
Chinese (zh)
Other versions
CN104729413A (en
Inventor
赵跃进
赵远
张雨旸
王浩文
常迎帅
赵昕
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Xian Jiaotong University
Original Assignee
Xian Jiaotong University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Xian Jiaotong University filed Critical Xian Jiaotong University
Priority to CN201510125641.4A priority Critical patent/CN104729413B/en
Publication of CN104729413A publication Critical patent/CN104729413A/en
Application granted granted Critical
Publication of CN104729413B publication Critical patent/CN104729413B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Landscapes

  • Length Measuring Devices By Optical Means (AREA)

Abstract

Puted the palms together before one product width online automatic detection and adjusting means and method the invention discloses one kind, the device includes the light source positioned at product bottom of putting the palms together before one and the video camera positioned at product top of putting the palms together before one, the output end of video camera is connected to the processor of image procossing, the output end of processor is connected with alarm controller, width adjusting device and the display for showing current detection product image and related data and state, the alarm of sound and light alarm can be sent by being connected with alarm controller, product image to be detected is input to processor by video camera, processor calculates product width, it can learn whether production process is normal by observing display, when abnormal, send alarm acousto-optic and control width adjusting device to carry out width adjustment.Detection time of the present invention is short, accuracy of detection is high, the assay intervals time is short, governing speed is fast, degree of regulation is high, it is possible to reduce wastes, improves product qualification rate, save a large amount of production costs and human resources.

Description

One kind is puted the palms together before one product width online automatic detection and adjusting means and method
Technical field
Puted the palms together before one Product checking technology the present invention relates to one kind, and in particular to one kind put the palms together before one product width online automatic detection with Adjusting means and method.
Background technology
The method of view-based access control model detection is widely used in steel production (strip, width of steel billet, thickness measure), wiring board system Make (automatic welding, soldered elements detection), (detection, non-defective unit and the defective products of various products size detect, weld seam matter for quality testing Amount detection) etc. various fields;Width Automatic adjustment method is widely used in the fields such as steel, tire tread, slab, control method There is the methods of PLC, stepper motor, PID.
It is a necessary production process of outer package product to put the palms together before one, and prints the enterprise of housing material, is being completed for printing Need to be puted the palms together before one packaging material afterwards, product given to manufacturing enterprise's (some needs are cut) after completion of putting the palms together before one, And, it is necessary to carry out process of putting the palms together before one in strict accordance with desired size during being puted the palms together before one, general allowable error is 0.2~ 0.5mm。
Product width of putting the palms together before one at present detection relies primarily on manual measurement and observation with regulation, and worker is by steel in production process Chi, micrometer or slide measure are close to product to measure, and whether observation measured value conforms to come the size for judging to put the palms together before one Ask, if undesirable, it is necessary to manual setting equipment, untill requirement is reached, this detection not only lost labor but also can not Lean on:1) this size detection often brings artificial defect to product, and the process of putting the palms together before one is continuous flow procedure, when worker is tight by steel ruler Patch product observation put the palms together before one size when, friction is had between steel ruler and product, thus cause product scuffing and cause waste product;2) hand middle-envelope bag-made machine Observed by people to manage, it is necessary to be spaced, general every 1000 meters measure three times, if hand middle-envelope bag-made machine speed of production is 60 ms/min Clock, then require worker's measurement in every 5 minutes once, 12 times/per hour, each production unit about measures 50 times, such measurement frequency Rate is that comparison is high for artificial;3) calculated three times according to every 1000 meters measurements, i.e., assay intervals are 3 meters, work as appearance During substandard product, the product that one-time detection is possible to waste is up to 300 meters, and this waste is huge;4) manual adjustment Speed is slow, and when the product width that detects to put the palms together before one is undesirable, staff first has to unclamp pressing plate, and adjustment press plate position arrives Suitable width, pressing plate is then tightened again, Primary regulation at least needs a few minutes, can cause larger waste;5) removed using hand middle-envelope bag-made machine Outside feeding, adjustment tension force, it is also necessary to which running check, which whether there is, to be scratched, scratch situations such as phenomenon and spread are uneven, it is desirable to Operator's strong sense of responsibility, notice are concentrated, but cause product matter of putting the palms together before one with reasons such as sense of responsibility because the observation of operator is horizontal Measure unqualified;6) in order to ensure the reliability of artificial detection, it is necessary to a large amount of high worker of Technical comparing, due to human resources and labor The reasons such as power cost are necessarily to bring cost to increase to manufacturing enterprise.
With the progress of science and technology, application image treatment technology, computer vision, machine vision technique carry out lossless inspection Survey more and more extensive, steel band, the width detection of steel billet, tobacco leaf detection, apple detection, boiler inspection such as in steel manufacture process Survey, glass detection, non-woven fabrics detection etc., wherein width detection is mainly used in width of steel band, rigid body crack width, printed circuit Line width etc..
Rely and build life et al. by adjusting light source chief ray with measurement plane into appropriate angle, obtain tested steel plate and background Reflection intensity distribution curve, realize that steel band surveys wide purpose.Tong Weiqi, Chen Baomo, Luo Xinbin are entered using double camera to strip Row measurement, the width of strip more accurate can be measured by demarcating, correcting;Leaf Cheng Long et al. is realized using laser range finder To the contactless width measure of strip;Ginger accounts for equality people realizes steel plate width detection using laser scanner technique and double CCD; Wang Zhi et al. realizes a kind of steel billet on-line detecting system using thermal infrared and digital image processing techniques;Zhou Genrong et al., grandson Yan Feng et al. application DSP or single-chip microcomputer and line array CCD realize that outside dimension measures;Qin Guang victory et al. application laser and CCD are realized A kind of measurement of steel plate width;Zhao Wei mono- et al. application line array CCDs and digital image processing techniques realize cut tobacco width measure; Nishiyama S et al. are studied the metric photography Measurement Algorithm of crack width;Luo Masiyang et al. are realized The lockhole Boundary Extraction and width measure of view-based access control model;Summer battalion's prestige et al. Appliance computer vision technology is to pipe tobacco width measure side Method is studied;The image segmentation algorithm surveyed in width online to hot-strip such as Li Yunxiang is studied;Jelen V. and Karch P. application image processing methods realize the measurement to continuous metal piece width;Ashida Takashi et al. are realized Detection and width measure to coke oven crack;Bear nation book et al. application image treatment technology realizes the measurement system of wiring board line width System;Prasanna Pratee et al. Appliance computer vision technologies realize the detection and analysis of deck crack;Zhang Jing Et al. applied statistics Process Control Theory high accuracy prints circuit board line width detection algorithm is studied;Sheng Xiaocheng application choosings Select detection zone, detection zone adds filling line, calculating the method for peak width of filling region covering, to realize figure wide Degree detection;The refined application of the hillside plot multiple lever elements parallel with paper width, by the contact and rotation that detect these light sensations Turn state and realize paper width detection;Shi Ou, Wang Wenqiang application infrared sensor realize the detection of width of paper money;History applies one The mechanical device such as fixed detection edge strip and spring bracing realizes the detection of flexible packaged battery core width[39];Middle paddy English Sticky position and the size of reference mark are compared in department's application realizes forming label and printing control;The big money text loyalties of Li Zhao pass through undertaking Being used cooperatively between the mechanical devices such as plate, datum plate, upper and lower limiting plate, bolt of stay axle and adjusting nut, realizes paper roll The detection of width;Field et al. realizes the control of hot-strip head and tail width using population genetic Optimization Algorithm;Huang Yi et al. Hot-strip control is studied using fuzzy;Horse Anxin et al. realizes tyre surface linkage using CCD Intellectualized Width Gauges Line width measures and speed regulation;Tight flat cutting edge of a knife or a sword et al., as master controller, utilizes special driving chip using Cortex-M3 L298N, design realize control and the detecting system of stepper motor.The system is gathered by electric current conversion chip MAX4172 and walked Stepper motor phase current, serial communication is carried out using RS-232 and host computer, the set of frequency and data for completing stepper motor are shown Etc. function;Zhu Yong devises a step motor control system based on PLC according to the application principle of stepper motor;Zu Yikang, It for controller, ULN2003 is stepper motor driver that Xu Miaojing, which uses STC89C52RC single-chip microcomputers, devise based on single-chip microcomputer and Visual Basic6.0 stepper motor tele-control system.The distinguishing feature of these research work is:1st, measured object belongs to firm Property material;2nd, measured object has light tight feature.Therefore, formed using the reflective of detected material, printing opacity, refractive nature Stripe shape, calculated using digital signal processing algorithm and artificial intelligence theory, so as to measure the width of tested object The information such as degree, height, thickness, reach the purpose of detection.
But product of putting the palms together before one is that material belongs to flexible material made of plastics mostly, has certain elasticity, in process of production Due to reasons such as machine vibration, delivery speed change, tension variations, shake is easily produced, such shake means when shooting Machine is fixed, the distance of subject to camera lens changes, and the change of this object distance causes the Target in Image size hair of collection Changing, and then cause the measurement data after image procossing to change and (measure width and mistake occur) so that width Detection difficulty increase.Product majority of putting the palms together before one is outer package product, and pattern form, color are various, the change of these shapes, color Change can bring difficulty to image procossing;In addition blank is left between product, because the randomness of IMAQ can cause blank Position changes at random, and then detection is impacted.Product width of putting the palms together before one testing requirements precision is high, therefore it is required that image resolution ratio Height, when image resolution ratio is high, lens edge distortion will influence accuracy of detection.Product width of putting the palms together before one automatically adjusts needs and adjusts speed Degree is fast, regulating time is short, degree of regulation is high.
The content of the invention
Puted the palms together before one product width online automatic detection and adjusting means and method it is an object of the invention to provide one kind, with gram The defects of above-mentioned prior art is present is taken, detection time of the present invention is short, accuracy of detection is high, the assay intervals time is short, governing speed It hurry up, degree of regulation height, it is possible to reduce waste, improve product qualification rate, save a large amount of production costs and human resources.
To reach above-mentioned purpose, the present invention adopts the following technical scheme that:
One kind is puted the palms together before one product width online automatic detection and adjusting means, including the light positioned at put the palms together before one product top or bottom Source and the video camera positioned at product top of putting the palms together before one, the output end of video camera are connected to the processor of image procossing, processor Output end be connected with alarm controller, width adjusting device and for show current detection product image and related data and The display of state, the alarm that can send sound and light alarm is connected with alarm controller.
Further, described light source has several, is distributed in the top or bottom of the product of putting the palms together before one of different production lines, institute The quantity for the video camera stated is identical with light source, respectively positioned at the product top of putting the palms together before one of different production lines, the output of each video camera End is connected with processor, and the output end of processor is connected with some alarm controllers and width adjusting for different production lines Device, the output end of processor are also associated with one and are used to show current different production line detection product images and related data With the display of state, the alarm that can send sound and light alarm is respectively connected with each alarm controller.
Further, width adjusting device includes the tune width controller being connected with the output end of processor, adjusts wide controller Motor is connected with by driver, positive and negative tooth screw mandrel is connected on motor, the orthodontic of positive and negative tooth screw mandrel is connected one respectively with anti-tooth The individual pressing plate for being used to adjust width.
A kind of detection and the adjusting method of put the palms together before one product width online automatic detection and adjusting means, comprise the following steps:
Step 1:The normal width for setting product of putting the palms together before one on a processor is w, it is allowed to which worst error value is v, and pixel ratio is g;
Step 2:Judge whether to start detection by processor, if so, then puted the palms together before one product image by camera acquisition, If it is not, then return to step 1;
Step 3:Step 2 acquired image is handled by processor;
Step 4:It is Dis that processor calculates product width of putting the palms together before one according to processing result image, and with the standard for product of putting the palms together before one Width compares;
Step 5:Whether processor exceeds error by the comparative result of step 4 judge the to put the palms together before one width of product, if so, Then send alarm go forward side by side line width regulation, subsequently into step 6, if it is not, being then directly entered step 6;
Step 6:Judge whether to stop detection by processor, if so, step 2 is then returned to, if it is not, then returning to step Three.
Further, image procossing includes binaryzation and eliminates burrs on edges in step 3.
Further, the method for binaryzation is as follows:If image size is:M*N;Pixel value is:F (x, y), wherein, 0≤x ≤M-1;0≤y≤N-1, then the value at (x, y) place be:
Pixel (x, y) place threshold value T (x, y) is:
T (x, y)=m (x, y)+k*s (x, y)
Neighborhood local window wherein centered on (x, y)Average m (x, y) be:
The standard deviation s (x, y) of neighborhood local window is:
Wherein neighborhood local window size is n*n, and n takes 15;Correction factor k takes -0.1.
Further, the method for eliminating burrs on edges is as follows:
1) with each pixel in a structural element scan image, covered with each pixel in structural element with it The pixel of lid does ' with ' computing, if being all ' 1 ', the pixel is ' 1 ', is otherwise ' 0 ';
2) with each pixel in a structural element scan image, covered with each pixel in structural element with it The pixel of lid does ' with ' computing, if being all ' 0 ', the pixel is ' 0 ', is otherwise ' 1 ';The burr of image border is eliminated, Wherein, structural element size is 3*3.
Further, the linear equation on two borders is found in step 3 using Hough transform algorithm, then Application way One or method two calculate width;
Method one:Width is calculated according to the slope-intercept form of linear equation:
The linear equation on one of border is:L1:Y=k1x+b1, the linear equation on another border is:L2:Y=k2x+ b2, k=(k1+k2)/2, then width
Method two:Width is calculated according to distance between beeline and dot formula:The linear equation on two borders is respectively L1:Y=k1x+ b1And L2:Y=k2x+b2, calculate L1Upper each point is to L2Distance, then calculate the average value of all distances, be to survey width:
The slope of two straight lines is respectively k1And k2, for L1On any point p (xi,yi), if L2It is upper point p (x to be presentj, yj), meet yj=yi+k1(xj-xi);Then p (xi,yi) and p (xj,yj) it is minimum range point pair, retain the point pair, and calculate and be somebody's turn to do The distance of point pair:If in the absence of the point for meeting above-mentioned condition, cut-off point p (xi,yi), average distance is: Wherein N represents total point logarithm, and this average distance is to be surveyed Width.
Further, the normal width comparative result of put the palms together before one product width and the product of putting the palms together before one that are calculated in step 4 For:U=| w-Dis |, the width of the product that judges to put the palms together before one in step 5 whether beyond the method for error be:If u >=v, production is represented Product width beyond allow error range, start alarm go forward side by side line width regulation;If u < v, represent that product width is allowing to miss Within the scope of difference, then into step 6.
Further, the method for width adjusting is:As Dis >=w, represent that current product width is wide, it is necessary to adjust narrow; As Dis < w, represent that current product width is narrow, it is necessary to adjust width, the distance of regulation is:Δ Dis=u/2.
Compared with prior art, the present invention has technique effect beneficial below:
Product image to be detected of putting the palms together before one is input to processor by the video camera of the present invention, and processor passes through a set of detection algorithm Processing, product width of putting the palms together before one is calculated, over the display, whether just worker can learn production process by observing display for display Often, when processor finds that product is abnormal, alarm acousto-optic is sent, while points out the error for mistake occur, and controls width to adjust Regulating device carries out width adjustment, and width adjustment, which is mainly realized, to be adjusted narrow (product width is wide), adjusts wide (product width is narrow), should Device can substitute artificial detection and regulation, and detection time is short, accuracy of detection is high, the assay intervals time is short, governing speed is fast, adjusts It is high to save precision, it is possible to reduce waste, improve product qualification rate, save a large amount of production costs and human resources.
Further, it is real by the way that the product width detection of putting the palms together before one of different production lines is connected into same processor with regulation Show shared a processor and display and realize a plurality of production line cooperation.
Brief description of the drawings
Fig. 1 is the device overall schematic of the present invention;
Fig. 2 is the assembling schematic diagram of a production line of the invention;
Fig. 3 is the assembling schematic diagram of a plurality of production line of the present invention;
Fig. 4 is double guide rail stabilization schematic diagrames of the present invention;
Fig. 5 is the four guide rail stabilization schematic diagrames of the present invention;
Fig. 6 is the width adjusting device structural representation of the present invention;
Fig. 7 is the workflow schematic diagram of the present invention.
Wherein, 1, camera lens;2nd, video camera;3rd, processor;4th, display;5th, alarm controller;6th, width adjusting device;7、 Alarm;8th, light source;9th, put the palms together before one product;10th, motor;11st, positive and negative tooth screw mandrel;12nd, pressing plate.
Embodiment
Embodiments of the present invention are described in further detail below in conjunction with the accompanying drawings:
Referring to Fig. 1 to Fig. 3, one kind is puted the palms together before one product width online automatic detection and adjusting means, including positioned at product 9 of putting the palms together before one The light source 8 and the video camera 2 positioned at the top of product 9 of putting the palms together before one of top or bottom, product 9 of putting the palms together before one are supported on light source 8 by guide shaft Portion or bottom, the guide shaft are horizontally disposed double guide shafts or four guide shafts that are divided to two rows to be arranged in parallel, the output end of video camera 2 Be connected to the processor 3 of image preprocessing, the output end of processor 3 be connected with alarm controller 5, width adjusting device 6 with And for showing the display 4 of current detection product image and related data and state, being connected with alarm controller 5 can The alarm 7 of sound and light alarm is sent, width adjusting device 6 includes the tune width controller being connected with the output end of processor 3, adjusts wide Controller is connected with motor 10 by driver, is connected to positive and negative tooth screw mandrel 11 on motor 10, the orthodontic of positive and negative tooth screw mandrel 11 with it is anti- A pressing plate 12 for being used to adjust width is connected on tooth respectively.
When controlling width detection and the regulation of a plurality of production line simultaneously, described light source 8 has several, is distributed in difference The top or bottom of the product 9 of putting the palms together before one of production line, the quantity of described video camera 2 is identical with light source 8, respectively positioned at different productions The top of product 9 of putting the palms together before one of line, the output end of each video camera 2 is connected with processor 3, if the output end of processor 3 is connected with Dry alarm controller 5 and width adjusting device 6 for different production lines, the output end of processor 3 are also associated with one and are used for The current different production lines detection product images of display and the display 4 of related data and state, on each alarm controller 5 The alarm 7 of sound and light alarm can be sent by being connected with.
Referring to Fig. 7, one kind is puted the palms together before one product width online automatic detection and adjusting method, is comprised the following steps:
Step 1:It is w that the normal width of product 9 of putting the palms together before one is set on processor 3, it is allowed to which worst error value is v, pixel ratio For g;
Step 2:Judge whether to start detection by processor 3, if so, then gathering the figure of product 9 of putting the palms together before one by video camera 2 Picture, if it is not, then returning to step 1;
Step 3:Step 2 acquired image is handled by processor 3, image procossing includes binaryzation and disappeared Except burrs on edges;
The method of binaryzation is as follows:If image size is:M*N;Pixel value is:F (x, y), wherein, 0≤x≤M-1;0≤y ≤ N-1, then the value at (x, y) place be:
The average and standard deviation of local window based on image calculate threshold value, and pixel (x, y) place threshold value T (x, y) is:
T (x, y)=m (x, y)+k*s (x, y)
Neighborhood local window wherein centered on (x, y)Average m (x, y) be:
The standard deviation s (x, y) of neighborhood local window is:
Wherein neighborhood local window size is n*n, and n takes 15;Correction factor k takes -0.1;
The method for eliminating burrs on edges is as follows:
1) with each pixel in a structural element (be usually 3*3 sizes) scan image, with structural element The pixel that each pixel covers with it does ' with ' computing, if being all ' 1 ', the pixel is ' 1 ', is otherwise ' 0 ';
2) with each pixel in a structural element (be usually 3*3 sizes) scan image, with structural element The pixel that each pixel covers with it does ' with ' computing, if being all ' 0 ', the pixel is ' 0 ', is otherwise ' 1 ';Eliminate The burr of image border;
Step 4:It is Dis that processor 3 calculates the width of product 9 of putting the palms together before one according to processing result image, and with product 9 of putting the palms together before one Normal width compares;Calculate width method be:The linear equation on two borders is found using Hough transform algorithm first, so Application way one or method two calculate width afterwards;
Method one:Width is calculated according to the slope-intercept form of linear equation:
The linear equation on one of border is:L1:Y=k1x+b1, the linear equation on another border is:L2:Y=k2x+ b2, k=(k1+k2)/2, then width
Method two:Width is calculated according to distance between beeline and dot formula:The linear equation on two borders is respectively L1:Y=k1x+ b1And L2:Y=k2x+b2, calculate L1Upper each point is to L2Distance, then calculate the average value of all distances, be to survey width:
The slope of two straight lines is respectively k1And k2, for L1On any point p (xi,yi), if L2It is upper point p (x to be presentj, yj), meet yj=yi+k1(xj-xi);Then p (xi,yi) and p (xj,yj) it is minimum range point pair, retain the point pair, and calculate and be somebody's turn to do The distance of point pair:If in the absence of the point for meeting above-mentioned condition, cut-off point p (xi,yi), average distance is: Wherein N represents total point logarithm, and this average distance is to be surveyed Width;
The normal width comparative result of the put the palms together before one width of product 9 and product 9 of putting the palms together before one that are calculated is:U=| w-Dis |;
Step 5:Whether processor 3 exceeds error by the comparative result of step 4 judge the to put the palms together before one width of product 9, if It is then to send alarm, subsequently into step 6, if it is not, being then directly entered step 6;
Specific method is:Judge to put the palms together before one in step 5 product 9 width whether beyond the method for error be:If u >=v, Represent product width beyond allow error range, start alarm go forward side by side line width regulation;If u < v, represent that product width exists Within the scope of allowable error, then into step 6;
Wherein, the method for width adjusting is:As Dis >=w, represent that current product width is wide, it is necessary to adjust narrow;When During Dis < w, represent that current product width is narrow, it is necessary to adjust width, the distance of regulation is:Δ Dis=u/2.
Step 6:Judge whether to stop detection by processor 3, if so, step 2 is then returned to, if it is not, then returning to step Three.
The implementation process of the present invention is described in further detail below:
It is online to design a set of satisfaction product width of putting the palms together before one using vision-based detection, image procossing, intellectual analysis principle by the present invention Automatic detection and the system of regulation, artificial detection and manually regulation, ginseng are substituted with Full automatic product width on-line checking and regulation See Fig. 2, data input cell, result display unit, regulation unit, alarm control unit form full-automatic on-line checking system System.As needed, can by multiple data input cells, alarm control unit, adjust wide one processor unit of units shared and As a result display unit realizes a plurality of production line cooperation, as shown in Figure 3.
In order to realize put the palms together before one product width online automatic detection and regulation, using video camera 2, camera lens 1, processor 3, display The compositions such as device 4, light source 8 automatically detect and regulating system, and video camera 2 and camera lens 1 are photoelectric conversion device, and will put the palms together before one product 9 The picture inversion formed in the presence of light is electric signal, and converts electrical signals to digital data transmission to processor 3.Processing Device 3 is the data processing unit of device, usually computer, and processor 3 receives the data of video camera 2, then at application image The calculating such as reason technology, mode identification technology, machine vision are puted the palms together before one the width of product 9, and judge whether product width meets the requirements, If undesirable, processor 3 sends alarm signal to alarm controller 5 and determines width adjusting direction and regulation step number, Then regulation is started.
Display 4 is used for showing current detection product image, shows related data and state (qualified, early warning, alarm), uses Make interactive interface, be easy to manual operation and observation.The signal of the reception processing device 3 of alarm controller 5, control alarm 7 send sound Light alarm, alarm 7 can send sound and light alarm signal, prompt operating personnel;Light source 8 is used for ensureing to put the palms together before one product 9 can be clear Processor 3 is presented to, can be irradiation light or transmitted light, guide rail is for reducing product shake, can be divided into four guide rails With double guide rails, width adjusting device 6 is used for adjusting the width of product 9 of putting the palms together before one, and by adjusting wide controller, driver, the grade of motor 10 forms, The signal of wide controller reception processing device 3 is adjusted, forms the control signal to driver after explaining the signal, driver controls herein The direction of controlled motor 10, the driving current rotated are produced in the presence of signal, motor 10 drives just under the driving of driving current Anti- tooth screw mandrel 11 moves, and reaches the purpose of width adjusting.
Light source 8 is positioned at the put the palms together before one top of product 9 or bottom, and video camera 2 is positioned at the put the palms together before one top of product 9, video camera 2 and processor 3 are connected, and product 9 image to be detected of putting the palms together before one is input to processor 3 by video camera 2, and processor 3 is by a set of detection algorithm Reason, calculates the width of product 9 of putting the palms together before one, is shown on display 4, worker can be by observing whether display 4 learns production process Normally, when processor 3 finds that product is abnormal, alarm acousto-optic is sent, while points out the error for mistake occur, and controls width Adjusting means 6 carries out width adjustment, and width adjustment, which is mainly realized, to be adjusted narrow (product width is wide), adjusts wide (product width mistake It is narrow).Processor 3 sends Regulate signal to wide controller is adjusted, and adjusts wide controller to send adjustment signal and adjustment according to adjustment needs Direction signal.Motor 10 drives executing agency's action in the case where adjusting the control of wide controller, and executing agency is a positive and negative tooth screw mandrel 11 two pressing plates 12 of drive move.
In process of production, product 9 is puted the palms together before one because tension force and other reasons can be shaken, and forms certain shake width Degree, it can so influence the precision of width detection.If in Fig. 4 in traditional support mode C guide shafts and B guide shafts distance near enough, A, B The material of the jitter of guide shaft is by the prevention of B guide shafts, and this shake will reduce or even disappearance;If in A guide shafts and B guide shafts Underface sets C guide shafts and D guide shafts respectively, forms structure as shown in Figure 5, can feed the right from the left side and discharge, can also be from The right charging left side discharging, the prevention from material when left side charging, the right discharging, shaken before A guide shafts by A, C guide shaft, It can reduce between C, D guide shaft and even eliminate shake, ensure the quality of collection image, can so be shaken with eliminating.

Claims (5)

  1. Product width online automatic detection and adjusting means 1. one kind is puted the palms together before one, it is characterised in that including on product of putting the palms together before one (9) The light source of portion or bottom (8) and the video camera (2) positioned at product of putting the palms together before one (9) top, product of putting the palms together before one (9) are supported on light by guide shaft Source (8) top or bottom, the guide shaft are used for reducing product of putting the palms together before one (9) shake, and the output end of video camera (2) is connected to image The processor (3) of processing, the output end of processor (3) are connected with alarm controller (5), width adjusting device (6) and are used for Current detection product image and the display of related data and state (4) are shown, is connected with and can send out on alarm controller (5) Go out the alarm (7) of sound and light alarm;
    Described light source (8) has several, is distributed in the top or bottom of the product of putting the palms together before one (9) of different production lines, described takes the photograph The quantity of camera (2) is identical with light source (8), respectively positioned at product of putting the palms together before one (9) top of different production lines, each video camera (2) Output end be connected with processor (3), the output end of processor (3) is connected with some controlling alarms for different production lines Device (5) and width adjusting device (6), the output end of processor (3) are also associated with one and are used to show current different production line inspections Product image and the display of related data and state (4) are surveyed, sound can be sent by being respectively connected with each alarm controller (5) The alarm (7) of light alarm;
    Width adjusting device (6) includes the tune width controller being connected with the output end of processor (3), adjusts wide controller to pass through driving Device is connected with motor (10), positive and negative tooth screw mandrel (11) is connected on motor (10), the orthodontic of positive and negative tooth screw mandrel (11) on anti-tooth with dividing Lian Jie not a pressing plate (12) that be used to adjust width.
  2. 2. detection and the adjusting method of put the palms together before one product width online automatic detection and the adjusting means described in claim 1, it is special Sign is, comprises the following steps:
    Step 1:It is w that the normal width of product (9) of putting the palms together before one is set on processor (3), it is allowed to which worst error value is v, pixel ratio For g;
    Step 2:Judge whether to start detection by processor (3), if so, then gathering product (9) of putting the palms together before one by video camera (2) Image, if it is not, then returning to step 1;
    Step 3:Step 2 acquired image is handled by processor (3), image procossing includes binaryzation and elimination Burrs on edges;
    The method of binaryzation is as follows:If image size is:M*N;Pixel value is:F (x, y), wherein, 0≤x≤M-1;0≤y≤N- 1, then the value at (x, y) place be:
    <mrow> <mi>b</mi> <mrow> <mo>(</mo> <mi>x</mi> <mo>,</mo> <mi>y</mi> <mo>)</mo> </mrow> <mo>=</mo> <mfenced open = "{" close = ""> <mtable> <mtr> <mtd> <mn>0</mn> </mtd> <mtd> <mrow> <mi>f</mi> <mrow> <mo>(</mo> <mi>x</mi> <mo>,</mo> <mi>y</mi> <mo>)</mo> </mrow> <mo>&lt;</mo> <mi>T</mi> <mrow> <mo>(</mo> <mi>x</mi> <mo>,</mo> <mi>y</mi> <mo>)</mo> </mrow> </mrow> </mtd> </mtr> <mtr> <mtd> <mn>255</mn> </mtd> <mtd> <mrow> <mi>f</mi> <mrow> <mo>(</mo> <mi>x</mi> <mo>,</mo> <mi>y</mi> <mo>)</mo> </mrow> <mo>&amp;GreaterEqual;</mo> <mi>T</mi> <mrow> <mo>(</mo> <mi>x</mi> <mo>,</mo> <mi>y</mi> <mo>)</mo> </mrow> </mrow> </mtd> </mtr> </mtable> </mfenced> </mrow>
    Pixel (x, y) place threshold value T (x, y) is:
    T (x, y)=m (x, y)+k*s (x, y)
    Neighborhood local window wherein centered on (x, y)Average m (x, y) be:
    <mrow> <mi>m</mi> <mrow> <mo>(</mo> <mi>x</mi> <mo>,</mo> <mi>y</mi> <mo>)</mo> </mrow> <mo>=</mo> <mfrac> <mn>1</mn> <msup> <mi>n</mi> <mn>2</mn> </msup> </mfrac> <munder> <mo>&amp;Sigma;</mo> <mrow> <mo>(</mo> <mi>x</mi> <mo>,</mo> <mi>y</mi> <mo>)</mo> <mo>&amp;Element;</mo> <msub> <mi>w</mi> <msup> <mi>x</mi> <mn>2</mn> </msup> </msub> </mrow> </munder> <mi>f</mi> <mrow> <mo>(</mo> <mi>x</mi> <mo>,</mo> <mi>y</mi> <mo>)</mo> </mrow> </mrow>
    The standard deviation s (x, y) of neighborhood local window is:
    <mrow> <mi>s</mi> <mrow> <mo>(</mo> <mi>x</mi> <mo>,</mo> <mi>y</mi> <mo>)</mo> </mrow> <mo>=</mo> <msqrt> <mrow> <mfrac> <mn>1</mn> <msup> <mi>n</mi> <mn>2</mn> </msup> </mfrac> <munder> <mo>&amp;Sigma;</mo> <mrow> <mo>(</mo> <mi>x</mi> <mo>,</mo> <mi>y</mi> <mo>)</mo> <mo>&amp;Element;</mo> <msub> <mi>w</mi> <msup> <mi>x</mi> <mn>2</mn> </msup> </msub> </mrow> </munder> <msup> <mi>f</mi> <mn>2</mn> </msup> <mrow> <mo>(</mo> <mi>x</mi> <mo>,</mo> <mi>y</mi> <mo>)</mo> </mrow> <mo>-</mo> <msup> <mi>m</mi> <mn>2</mn> </msup> <mrow> <mo>(</mo> <mi>x</mi> <mo>,</mo> <mi>y</mi> <mo>)</mo> </mrow> </mrow> </msqrt> </mrow>
    Wherein neighborhood local window size is n*n, and n takes 15;Correction factor k takes -0.1;
    The method for eliminating burrs on edges is as follows:
    1) with each pixel in a structural element scan image, covered with each pixel in structural element with it Pixel does ' with ' computing, if being all ' 1 ', the pixel is ' 1 ', is otherwise ' 0 ';
    2) with each pixel in a structural element scan image, covered with each pixel in structural element with it Pixel does ' with ' computing, if being all ' 0 ', the pixel is ' 0 ', is otherwise ' 1 ';The burr of image border is eliminated, wherein, Structural element size is 3*3;
    Step 4:It is Dis that processor (3) calculates product (9) width of putting the palms together before one according to processing result image, and with product of putting the palms together before one (9) Normal width compare;
    Step 5:Whether processor (3) exceeds error by the comparative result of step 4 judge the to put the palms together before one width of product (9), if Be then send alarm go forward side by side line width regulation, subsequently into step 6, if it is not, being then directly entered step 6;
    Step 6:Judge whether to stop detection by processor (3), if so, step 2 is then returned to, if it is not, then returning to step 3.
  3. 3. detection according to claim 2 and adjusting method, it is characterised in that Hough transform algorithm is used in step 3 The linear equation on two borders is found, then Application way one or method two calculate width;
    Method one:Width is calculated according to the slope-intercept form of linear equation:
    The linear equation on one of border is:L1:Y=k1x+b1, the linear equation on another border is:L2:Y=k2x+b2, k =(k1+k2)/2, then width
    Method two:Width is calculated according to distance between beeline and dot formula:The linear equation on two borders is respectively L1:Y=k1x+b1With L2:Y=k2x+b2, calculate L1Upper each point is to L2Distance, then calculate the average value of all distances, be to survey width:
    The slope of two straight lines is respectively k1And k2, for L1On any point p (xi,yi), if L2It is upper point p (x to be presentj,yj), Meet yj=yi+k1(xj-xi);Then p (xi,yi) and p (xj,yj) it is minimum range point pair, retain the point pair, and calculate the point pair Distance:If in the absence of the point for meeting above-mentioned condition, cut-off point p (xi, yi), average distance is: Wherein N represents total point logarithm, and this average distance is to survey width.
  4. 4. detection according to claim 2 and adjusting method, it is characterised in that the product of putting the palms together before one being calculated in step 4 (9) the normal width comparative result of width and product of putting the palms together before one (9) is:U=| w-Dis |, product (9) of putting the palms together before one is judged in step 5 Width whether beyond the method for error be:If u >=v, represent that product width beyond the error range allowed, starts alarm and gone forward side by side Line width is adjusted;If u < v, product width is represented within the scope of allowable error, then into step 6.
  5. 5. detection according to claim 2 and adjusting method, it is characterised in that the method for width adjusting is:As Dis >=w When, represent that current product width is wide, it is necessary to adjust narrow;As Dis < w, represent that current product width is narrow, it is necessary to adjust Width, the distance of regulation are:Δ Dis=u/2.
CN201510125641.4A 2015-03-20 2015-03-20 One kind is puted the palms together before one product width online automatic detection and adjusting means and method Active CN104729413B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201510125641.4A CN104729413B (en) 2015-03-20 2015-03-20 One kind is puted the palms together before one product width online automatic detection and adjusting means and method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201510125641.4A CN104729413B (en) 2015-03-20 2015-03-20 One kind is puted the palms together before one product width online automatic detection and adjusting means and method

Publications (2)

Publication Number Publication Date
CN104729413A CN104729413A (en) 2015-06-24
CN104729413B true CN104729413B (en) 2017-12-08

Family

ID=53453541

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201510125641.4A Active CN104729413B (en) 2015-03-20 2015-03-20 One kind is puted the palms together before one product width online automatic detection and adjusting means and method

Country Status (1)

Country Link
CN (1) CN104729413B (en)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106370112A (en) * 2016-11-01 2017-02-01 上海兰宝传感科技股份有限公司 Universal width detection system
CN111761829A (en) * 2020-06-24 2020-10-13 深圳市博泰数码智能技术有限公司 Width adjusting method applied to palm folding machine

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1680072A (en) * 2004-04-08 2005-10-12 电子科技大学 Precisive measurement of static knife profile
CN101216287A (en) * 2008-01-11 2008-07-09 天津大学 Spring concentricity and verticality measuring meter and its measurement method
CN102252615A (en) * 2010-05-19 2011-11-23 北京京东方光电科技有限公司 Stamped mark width measuring device and method

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPS60122304A (en) * 1983-11-09 1985-06-29 Shinetsu Eng Kk Automatic size measuring device

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1680072A (en) * 2004-04-08 2005-10-12 电子科技大学 Precisive measurement of static knife profile
CN101216287A (en) * 2008-01-11 2008-07-09 天津大学 Spring concentricity and verticality measuring meter and its measurement method
CN102252615A (en) * 2010-05-19 2011-11-23 北京京东方光电科技有限公司 Stamped mark width measuring device and method

Also Published As

Publication number Publication date
CN104729413A (en) 2015-06-24

Similar Documents

Publication Publication Date Title
CN113934190B (en) Corrugated board production quality regulation and control method based on machine vision
CN104713486B (en) One kind is puted the palms together before one product width on-line measuring device and method
CN109807076B (en) Automatic optical detection method and equipment for pressure-sensitive double-sided adhesive tape
CN103940350A (en) Coating-machine online pole plate thickness measurement device and thickness measurement adjustment method
CN110648345B (en) Method and system for detecting flow of tobacco shred materials on conveying belt based on light field imaging
CN104729413B (en) One kind is puted the palms together before one product width online automatic detection and adjusting means and method
CN206131746U (en) Be applied to steel rolling production line&#39;s heating furnace loading attachment
AU2022349296A1 (en) Road surface technical condition detection method and device based on three-dimensional contour
CN110376211B (en) Wet-process-gummed synthetic leather hemming on-line detection device and method
EP0138430A2 (en) Process and apparatus for strip flatness and tension measurements
CN102581244A (en) Online control system and control method for surface quality of continuous casting billet
CN115631138A (en) Zirconium alloy plate laser cutting quality monitoring method and device
CN105301008A (en) Tipping paper quality test warning system and on-line quality test method
US6909502B2 (en) Method and apparatus for measuring ripple and distortion in a transparent material
CN115384052A (en) Intelligent laminating machine automatic control system
CN203518957U (en) Detection system of laser scribing morphology
CN212779128U (en) Steel plate contour measuring device
DE102018202618A1 (en) A grinding apparatus and method for detecting a condition of a workpiece machined by a grinding apparatus
CN109622926B (en) Ingot block edge detection method for ingot casting process
CN213903347U (en) Paster apparent on-line measuring equipment
CN111709325B (en) Automatic irradiation method for hot rolling heating furnace slab
CA3076342A1 (en) Aligning sensors on vehicles using sensor output
CN108387585B (en) Frequency automatic adjustment type printing picture stroboscopic detection device and detection method
CN114210741B (en) Strip steel surface roughness online detection device and method
CN215449986U (en) Outdoor intelligent inspection device for wind power plant

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant