CN103499297A - CCD (Charge Coupled Device)-based high-accuracy measuring method - Google Patents

CCD (Charge Coupled Device)-based high-accuracy measuring method Download PDF

Info

Publication number
CN103499297A
CN103499297A CN201310509532.3A CN201310509532A CN103499297A CN 103499297 A CN103499297 A CN 103499297A CN 201310509532 A CN201310509532 A CN 201310509532A CN 103499297 A CN103499297 A CN 103499297A
Authority
CN
China
Prior art keywords
image
point
measured
vector
template
Prior art date
Application number
CN201310509532.3A
Other languages
Chinese (zh)
Other versions
CN103499297B (en
Inventor
耿磊
王忠强
Original Assignee
爱科维申科技(天津)有限公司
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 爱科维申科技(天津)有限公司 filed Critical 爱科维申科技(天津)有限公司
Priority to CN201310509532.3A priority Critical patent/CN103499297B/en
Publication of CN103499297A publication Critical patent/CN103499297A/en
Application granted granted Critical
Publication of CN103499297B publication Critical patent/CN103499297B/en

Links

Abstract

The invention belongs to the field of vision, relates to a CCD (Charge Coupled Device)-based high-accuracy measuring method, and breaks through the limitation that the conventional CCD-based size measuring technology is mostly applied to measurement of small-sized objects. An algorithm comprises the following steps: building a camera imaging model, calibrating internal and external parameters of a camera, and correcting images according to a distortion model and internal parameter data; putting forward a method for distinguishing the edges of the upper and lower surfaces of a tested object by using neighborhood information, extracting the sub-pixel edge information of the upper surface of the tested object by using a Canny filter, building a template according to CAD (Computer Aided Design) data of a detected part, accurately locating and identifying the tested object by using a shape-based template matching algorithm, acquiring the depth information of the tested object according to a dot laser, and constraining and converting the posture of the upper surface of the tested object, wherein a profile distance is taken as the machining error measure of the tested object. The algorithm can be applied to size measurement of large-sized planar parts, and the accuracy can reach a sub-pixel grade.

Description

A kind of high-precision measuring method based on CCD
Technical field
The invention belongs to technical field of image processing, relate to a kind of high-acruracy survey side based on CCD, can provide high precision, high efficiency detection for mechanical component processing.
Background technology
Modern industry is just towards the future development of process automation, high speed, precise treatment, and enterprise is more and more higher to the requirement on machining accuracy of product.In mechanical processing process, because lathe vibrations, tool wear, tool setting skew various reasons all can be directly or the remote effect machining precision.The important channel that obtains the workpiece size machining precision is to measure.Measurement is the key link in product manufacture, is the Main Means of judgement production quality.The development of advanced manufacturing technology, proposed high precision, high-level efficiency, good flexibility requirement to measurement, also more and more emphasizes the real-time online non-cpntact measurement.For example the sheet metal precision cuts enterprise when buying and use high precision to cut instrument, how to understand the precision of cropping of products, this is long-term puzzled many slip-stick artists in the industry and operating personnel's a difficult problem, especially, when cutting irregularly shaped part, is difficult to especially do high-precision measurement.
Increasing researchist and enterprise project the CCD size measurement technique to sight, it has become a kind of very effective non-contact detecting technology, processing, detection and control procedure are combined together, can meet that measuring speed is fast, precision is high, the requirement of noncontact and dynamic auto measurement.The CCD size measurement technique is more more advantageous than existing mechanical type, optical profile type, electromagnetism formula weight instrument, detect the application in identification at miniature, large-scale, complicated, many curve surface work pieces especially, effectively avoided manual detection identification labour intensity large, efficiency is low, fatigability and traditional detection function ratio are more single, robotization, the not high shortcoming of intelligent degree, and accuracy of identification and real-time are also become better and better.But the system that the current domestic CCD of utilization carries out industrial real-time online detection is few, and multiplex line array CCD, precision is not high, requires individually the high-precision system technology that adopt a plurality of line array CCDs splicings more, has by many low resolution area array CCDs yet and splices to reach high-precision requirement.
As can be seen here, CCD can be used to carry out dimensional measurement, workpiece location and profile and aims at, but be subject to the impact of site environment, light source, systemic resolution and imaging precision, apply single area array CCD and carry out the high-accuracy large-scale measurement, precision reaches sub-pixel, and drops into the automatic assembly line of industrial actual motion, the domestic report that there is not yet, though some theoretical researches are arranged, all are not enough to solve existing application area array CCD and carry out the existing problem of large-size images measuring system.The high-acruracy survey (precision is higher than 0.05mm) of large sized object (size surpasses 500 mm * 500mm) faces very large difficulty.Therefore developing a kind of CCD of utilization carries out the accessory size automatic measurement system, realize the quick high accuracy measurement of large-scale part, reduce the personal error of bringing in measuring process, to increasing economic efficiency, the automaticity of raising system and intelligent degree, significant.
Summary of the invention
The objective of the invention is to overcome the above-mentioned deficiency of prior art, high-precision measuring method based on CCD has been proposed, gordian technique in the step of utilizing this method to provide, make every effort to for mechanical component processing provides high precision, high efficiency detection, to avoid Traditional Man to detect the intrinsic shortcoming that subjectivity, fatiguability, speed are slow, cost is high, intensity is large.Especially for the support provided on theory and technology is provided the sheet metal precision, make to cut the part detection and there is consistance, accuracy and repeatability.
The present invention realizes by following method: the high-precision measuring method based on CCD, it is characterized in that, and comprise the following steps:
(1) calibrating camera inside and outside parameter correcting image;
(2) location measured object;
(3) distinguish lower limb on measured object, extract the upper surface sub-pixel edge;
(4) measure the measured object height, determine the edge actual coordinate of measured object;
(5) import the cad data of tested part, drawing template establishment coupling;
(6) template and measured object edge model compared and carry out error analysis.
Substantive distinguishing features of the present invention is, set up a mechanical component detection model based on machine vision, and this model generalization has used camera calibration, sub-pixel edge profile to extract and the template matching technique based on shape.At first, according to camera imaging model, camera is carried out to the internal and external parameter demarcation, according to the correcting distorted image of inner parameter.Then, the location measured object, distinguish lower limb it on and extract the sub-pixel edge of upper surface, imports the cad data model, adopts the template matching method based on shape to mate measured object.The last tolerance using profile distance as error.The present invention has following advantage compared with prior art:
1. propose to utilize dot laser to obtain the depth information of measured object, for the attitude conversion of measurement plane provides prior imformation, thereby realized the high-acruracy survey of planar metal sheet material, substrate different-thickness measured object.
2. the present invention adopts the edge extracting of sub-pix when edge calculation, and precision is higher.
3. proposed to distinguish with neighborhood information the method at measured object upper and lower surface edge.
The accompanying drawing explanation
Fig. 1: measuring method process flow diagram of the present invention;
Fig. 2: measuring system mount scheme schematic diagram;
Fig. 3: dimensional measurement thickness approach key diagram.
In figure: 1. video camera, 2. measured object, 3. light source, 4. standard gauge block.
Embodiment
Process flow diagram of the present invention as shown in Figure 1, at first video camera carries out the inside and outside parameter demarcation, utilize the data of demarcating to carry out distortion correction to the measured object image, then locate measured object, distinguish measured object upper and lower surface edge and extract the sub-pixel edge of upper surface, afterwards according to the cad data drawing template establishment, the accurate fixation and recognition measured object of the template matching algorithm of utilization based on shape, obtain the measured object depth information according to dot laser, constraint conversion measured object upper surface attitude, compare template and measured object and carry out error analysis.Below in conjunction with accompanying drawing, the specific implementation process of technical solution of the present invention is illustrated.
1. system structure design
System structure design as shown in Figure 2.Wherein, for the parts profile of high definition is provided, backlight adopts high density LED array face that the high intensity backlight illumination is provided, light irradiates from the measured object rear by light guide plate and diffuser plate, the CCD camera is arranged on the measurement plane top, optical axis is perpendicular to measurement plane, and dot laser is positioned over measurement plane one side.
2. camera interior and exterior parameter is demarcated and correcting image
Set up the nonlinear imaging geometric model of area array camera, based on two-step approach thought, camera is demarcated, obtain inner parameter: the optics of camera and geometric parameter, and external parameter: the position orientation relation between camera coordinate system and world coordinate system.According to distortion model and internal reference data correcting plane part image.
3. location measured object
In order to dwindle matching range, improve speed and the precision of coupling, at first need measured object is located.Owing to there being the gray scale difference of highly significant between measured object and image background, adopt the histogrammic dividing method of intensity-based, after the histogram Gaussian smoothing to image, selecting the peak-to-peak minimum value of histogram two is that threshold value is cut apart, the zone obtained is with 3 3 square structure element expands, and the zone arranged after expansion is area-of-interest (ROI).
4. distinguish the upper and lower surface edge, extract the sub-pixel edge of upper surface
Certain thickness and size are arranged when larger at measured object, have lower limb when distance camera one side, and, when the camera opposite side, only have coboundary, so marginal information need to distinguish coboundary and lower limb, and only extract the sub-pixel edge of measured object upper surface.
The method of employing based on 16 neighborhoods distinguished upper lower limb, to the arbitrary pixel p in ROI, calculates the outer minimum gray value of 16 neighborhoods if minimum value is greater than threshold value T, think that this pixel does not belong to the pixel of measured object upper surface, it is made as to 255, computing method are: establish set and establish set
????????????????????????????(1)
???????????????????????????????????????????????????(2)
Wherein for the gradation of image value;
mfor the set of image array ordinate;
nfor the set of image array ordinate;
afor the set of image array horizontal ordinate;
bfor the set of image array horizontal ordinate;
rfor the image array ordinate;
cfor the image array horizontal ordinate;
tfor the threshold value of setting;
The place of image border point in distributed function slope maximum.After the one dimension Continuous Gray Scale distribution function in Recovery image edge transition zone, sub-pixel edge is positioned at the extreme point position of distributed function slope.Use the Canny wave filter to obtain high-precision sub-pixel edge.Paper " A Computational Approach to Edge Detection " J. Canny in IEEE transactions on Pattern Analysis and Machine Intelligence, Volume 8, Issue 6, and pp. 679-698 (ISSN:0162-8828) has described this edge extracting technology.
5. obtain the measured object elevation information, determine the actual edge coordinate of measured object
Adopt dot laser to measure the measured object height.During measurement, at first laser instrument projects point-like laser to the measured object surface, and on the note image, the point-like laser center is , then place known precise height gauge block to plane, project point-like laser on gauge block and pictures taken, on the note image, the point-like laser center is , then point-like laser is projected on the measured object surface, clapped the imaging of penetrating down on body surface now by ccd video camera, on the note image, the point-like laser center is , because testee is different from the gauge block height, so twice imaging is different in the horizontal direction, thereby utilize trigonometric calculations to go out object thickness can be tried to achieve by formula (3) (4):
????????????????????????????(3)
?????????????????????????(4)
In formula:
hthickness for measured object;
dit is the distance between 2;
rfor the image array ordinate;
cfor the image array horizontal ordinate;
After recording the thickness of measured object, according to video camera external parameter and thickness information, measurement plane is transformed into to the upper plane of measured object, obtains the coordinate of measured object coboundary under world coordinate system.
6. import the cad data of tested part, drawing template establishment coupling
Adopted a kind of matching algorithm based on shape, can effectively solve that target rotates, the coupling of the image of translation.Concrete grammar is: at first, according to the cad data drawing template establishment of measured object, each puts associated direction vector the CALCULATION CAD image border.Next establishes the point set of template image , the direction vector of each point association ; The template image central point is P, and the direction vector of each point of image to be checked is ; During registration, the calculation template image center ppoint to image to be checked transformation matrix a, by affined transformation, template image is pressed to transformation matrix aintegral translation, the template image point set after being converted, be designated as , wherein the direction vector after the conversion is ; All in last computational transformation rear pattern plate direction vector with image corresponding point direction vector to be checked the summation of point set, the summation of this dot product is exactly similarity measure s:
??????????????????????????(5)
sfor the similarity value;
direction vector for each point on template image;
direction vector for each point on image to be checked;
for vector horizontal ordinate;
for vector ordinate;
X is vector horizontal ordinate;
Y is vector ordinate;
When the similarity value swhile reaching user-defined threshold value, just think at point found the example be complementary with template.
7. error analysis
If template image marginal point coordinate is set , image border to be detected point coordinate is set ; Departure between two image borders is defined as for, computing method are: at first, to each point in A, calculate its on B near distance a little like Euclidean distance, the distance symbol mean, computing method are suc as formula (8); Then, the distance obtained is sorted, got the departure of its middle distance minimum value for this point, used mean, last, calculate set be the departure between two image borders ;
?????????????????????????????(6)
??????????????????????????????????(7)
????????????????(8)
it is the departure between two image border coordinate sets;
Be 1 a to another set B minor increment a little;
A is the image border point;
D is the distance between 2;
Can calculate the error of part thus: traversal , according to the tolerance setting threshold ; When perhaps the time, the illustrated planar part do not meet tolerance herein, and be deviate.
In sum, the present invention can differentiate between images middle plateform part the upper and lower surface edge, extract the upper surface sub-pixel edge, the video camera that utilizes demarcation to obtain is joined acquisition part edge physical size outward, after completing images match, part physical size and cad model are compared, obtain detecting the mismachining tolerance value of flat part.Compare the Traditional Man detection method, the method has great improvement on the speed of flat part quality testing and precision.
According to the above description, can realize the solution of the present invention in conjunction with art technology.

Claims (6)

1. the high-precision measuring method based on CCD, comprise the following steps:
Calibrating camera inside and outside parameter correcting image;
The location measured object;
Distinguish lower limb on measured object, extract the upper surface sub-pixel edge;
Measure the measured object height, determine the edge actual coordinate of measured object;
Import the cad data of tested part, drawing template establishment coupling;
Template and measured object edge model are compared and carry out error analysis.
2. a kind of high-acruracy survey algorithm based on CCD according to claim 1, it is characterized in that, in step (2), in order to dwindle matching range, improve speed and the precision of coupling, at first need measured object is located, owing to there being the gray scale difference of highly significant between measured object and image background, adopt the histogrammic dividing method of intensity-based, after the histogram Gaussian smoothing to image, selecting the peak-to-peak minimum value of histogram two is that threshold value is cut apart, and the zone obtained is with 3 3 square structure element expands, and the zone arranged after expansion is region of interest ROI.
3. a kind of high-acruracy survey algorithm based on CCD according to claim 1, is characterized in that, in step (3), adopts the method based on 16 neighborhoods to distinguish upper lower limb, to the arbitrary pixel p in ROI, calculates the outer minimum gray value of 16 neighborhoods if minimum value is greater than threshold value T, think that this pixel does not belong to the pixel of measured object upper surface, its value is made as to 255, computing method are: establish set
????????????????????????????(1)
???????????????????????????????????????????????????(2)
Wherein for the gradation of image value;
mfor the set of image array ordinate;
nfor the set of image array ordinate;
afor the set of image array horizontal ordinate;
bfor the set of image array horizontal ordinate;
rfor the image array ordinate;
cfor the image array horizontal ordinate;
tfor the threshold value of setting;
The place of image border point in distributed function slope maximum, after the one dimension Continuous Gray Scale distribution function in Recovery image edge transition zone, sub-pixel edge is positioned at the extreme point position of distributed function slope, uses the Canny wave filter to obtain high-precision sub-pixel edge.
4. a kind of high-acruracy survey algorithm based on CCD according to claim 1, is characterized in that, in step (4), adopt dot laser to measure the measured object height, during measurement, at first laser instrument projects point-like laser to the measured object surface, and on the note image, the point-like laser center is , then place known precise height gauge block to plane, project point-like laser on gauge block and pictures taken, on the note image, the point-like laser center is , then point-like laser is projected on the measured object surface, clapped the imaging of penetrating down on body surface now by ccd video camera, on the note image, the point-like laser center is , because testee is different from the gauge block height, so twice imaging is different in the horizontal direction, thereby utilize trigonometric calculations to go out object thickness can be tried to achieve by formula (3), (4):
????????????????????????????(3)
?????????????????????????(4)
In formula:
hthickness for measured object;
dit is the distance between 2;
rfor the image array ordinate;
cfor the image array horizontal ordinate;
After recording the thickness of measured object, according to video camera external parameter and thickness information, measurement plane is transformed into to the upper plane of measured object, obtains the coordinate of measured object coboundary under world coordinate system.
5. a kind of high-acruracy survey algorithm based on CCD according to claim 1, is characterized in that, in step (5), adopted a kind of matching algorithm based on shape, effectively solves that target rotates, the coupling of the image of translation; Concrete grammar is: at first, according to the cad data drawing template establishment of measured object, each puts associated direction vector the CALCULATION CAD image border; Next establishes the point set of template image , the direction vector of each point association ; The template image central point is P, and the direction vector of each point of image to be checked is ; During registration, the calculation template image center ppoint to image to be checked transformation matrix a, by affined transformation, template image is pressed to transformation matrix aintegral translation, the template image point set after being converted, be designated as , wherein the direction vector after the conversion is ; All in last computational transformation rear pattern plate direction vector with image corresponding point direction vector to be checked the summation of point set, the summation of this dot product is exactly similarity measure s:
??????????????????????????(5)
sfor the similarity value;
direction vector for each point on template image;
direction vector for each point on image to be checked;
for vector horizontal ordinate;
for vector ordinate;
X is vector horizontal ordinate;
Y is vector ordinate;
When the similarity value swhile reaching user-defined threshold value, just think at point found the example be complementary with template.
6. a kind of high-acruracy survey algorithm based on CCD according to claim 1, is characterized in that, in step (6),
If template image marginal point coordinate is set , image border to be detected point coordinate is set ; Departure between two image borders is defined as for, computing method are: at first, to each point in A, calculate its on B near distance a little like Euclidean distance, the distance symbol mean, computing method are suc as formula (8); Then, the distance obtained is sorted, got the departure of its middle distance minimum value for this point, used mean, last, calculate set be the departure between two image borders ;
?????????????????????????????(6)
??????????????????????????????????(7)
????????????????(8)
it is the departure between two image border coordinate sets;
Be 1 a to another set B minor increment a little;
A is the image border point;
D is the distance between 2;
Can calculate the error of part thus: traversal , according to the tolerance setting threshold ; When perhaps the time, the illustrated planar part do not meet tolerance herein, and be deviate.
CN201310509532.3A 2013-10-25 2013-10-25 A kind of high-precision measuring method based on CCD CN103499297B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201310509532.3A CN103499297B (en) 2013-10-25 2013-10-25 A kind of high-precision measuring method based on CCD

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201310509532.3A CN103499297B (en) 2013-10-25 2013-10-25 A kind of high-precision measuring method based on CCD

Publications (2)

Publication Number Publication Date
CN103499297A true CN103499297A (en) 2014-01-08
CN103499297B CN103499297B (en) 2016-01-13

Family

ID=49864528

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201310509532.3A CN103499297B (en) 2013-10-25 2013-10-25 A kind of high-precision measuring method based on CCD

Country Status (1)

Country Link
CN (1) CN103499297B (en)

Cited By (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103776378A (en) * 2014-02-27 2014-05-07 上海思琢自动化科技有限公司 Non-contact type flexible on-line dimension measurement system
CN104075666A (en) * 2013-03-28 2014-10-01 株式会社三丰 Enhanced edge detection tool for edges of irregular surfaces
CN104228049A (en) * 2014-09-17 2014-12-24 西安交通大学 Machine vision based online blow molding product measuring method
CN104359403A (en) * 2014-11-21 2015-02-18 天津工业大学 Plane part size measurement method based on sub-pixel edge algorithm
CN104457577A (en) * 2014-12-19 2015-03-25 上海工业自动化仪表研究院 Machine-vision-oriented non-contact type workpiece positioning and measuring method
CN104655018A (en) * 2015-03-16 2015-05-27 武汉大学 System and method for detecting end surface size of spline housing
CN104923593A (en) * 2015-05-20 2015-09-23 南京航空航天大学 Vision-based positioning method for top layer bending plate
CN106041937A (en) * 2016-08-16 2016-10-26 河南埃尔森智能科技有限公司 Control method of manipulator grabbing control system based on binocular stereoscopic vision
TWI577493B (en) * 2014-12-26 2017-04-11 財團法人工業技術研究院 Calibration method and automatic apparatus using the same
CN106643483A (en) * 2016-09-28 2017-05-10 宁波舜宇智能科技有限公司 work piece detection method and device
CN107514974A (en) * 2017-09-07 2017-12-26 唐冬香 A kind of method and system of lathe detection workpiece
CN107705293A (en) * 2017-09-14 2018-02-16 广东工业大学 A kind of hardware dimension measurement method based on CCD area array cameras vision-based detections
CN107767369A (en) * 2017-09-27 2018-03-06 杭州迈锐钶科技有限公司 A kind of the defects of buret detection method and device
CN108871185A (en) * 2018-05-10 2018-11-23 苏州大学 Method, apparatus, equipment and the computer readable storage medium of piece test
CN109211136A (en) * 2018-08-31 2019-01-15 广州大学 A kind of Watch glass cover board profile tolerance detection method
CN109990707A (en) * 2019-04-02 2019-07-09 天津工业大学 A kind of cut-parts eye knife detection method based on corner constraint
CN111174703A (en) * 2020-02-24 2020-05-19 湖南工业大学 Non-contact size measurement method based on machine vision

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH118798A (en) * 1997-06-17 1999-01-12 Nok Corp Ccd camera system and image processing method
JP2001041725A (en) * 1999-07-28 2001-02-16 Nikon Corp Shape measurement device
CN101144703A (en) * 2007-10-15 2008-03-19 陕西科技大学 Article geometrical size measuring device and method based on multi-source image fusion
CN101576372A (en) * 2009-04-30 2009-11-11 上海理工大学 Automatic detection device for size of used position of surgical instrument and detection method thereof
CN102901444A (en) * 2012-08-29 2013-01-30 浙江大学 Method for detecting component size based on matching pursuit (MP) wavelet filtering and detecting system thereof

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH118798A (en) * 1997-06-17 1999-01-12 Nok Corp Ccd camera system and image processing method
JP2001041725A (en) * 1999-07-28 2001-02-16 Nikon Corp Shape measurement device
CN101144703A (en) * 2007-10-15 2008-03-19 陕西科技大学 Article geometrical size measuring device and method based on multi-source image fusion
CN101576372A (en) * 2009-04-30 2009-11-11 上海理工大学 Automatic detection device for size of used position of surgical instrument and detection method thereof
CN102901444A (en) * 2012-08-29 2013-01-30 浙江大学 Method for detecting component size based on matching pursuit (MP) wavelet filtering and detecting system thereof

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
陈旸: "基于面阵CCD的工业零部件典型尺寸检测系统研究", 《中国优秀硕士论文全文》 *

Cited By (24)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104075666A (en) * 2013-03-28 2014-10-01 株式会社三丰 Enhanced edge detection tool for edges of irregular surfaces
CN104075666B (en) * 2013-03-28 2017-08-29 株式会社三丰 Enhanced edge detection tool for irregular surface edge
CN103776378A (en) * 2014-02-27 2014-05-07 上海思琢自动化科技有限公司 Non-contact type flexible on-line dimension measurement system
CN104228049A (en) * 2014-09-17 2014-12-24 西安交通大学 Machine vision based online blow molding product measuring method
CN104359403A (en) * 2014-11-21 2015-02-18 天津工业大学 Plane part size measurement method based on sub-pixel edge algorithm
CN104359403B (en) * 2014-11-21 2017-03-29 天津工业大学 Planar part dimension measurement method based on sub-pixel edge algorithm
CN104457577A (en) * 2014-12-19 2015-03-25 上海工业自动化仪表研究院 Machine-vision-oriented non-contact type workpiece positioning and measuring method
TWI577493B (en) * 2014-12-26 2017-04-11 財團法人工業技術研究院 Calibration method and automatic apparatus using the same
US10209698B2 (en) 2014-12-26 2019-02-19 Industrial Technology Research Institute Calibration method and automation machining apparatus using the same
CN104655018A (en) * 2015-03-16 2015-05-27 武汉大学 System and method for detecting end surface size of spline housing
CN104655018B (en) * 2015-03-16 2017-07-25 武汉大学 A kind of spline housing face size detecting system and method
CN104923593A (en) * 2015-05-20 2015-09-23 南京航空航天大学 Vision-based positioning method for top layer bending plate
CN104923593B (en) * 2015-05-20 2017-01-11 南京航空航天大学 Vision-based positioning method for top layer bending plate
CN106041937A (en) * 2016-08-16 2016-10-26 河南埃尔森智能科技有限公司 Control method of manipulator grabbing control system based on binocular stereoscopic vision
CN106643483A (en) * 2016-09-28 2017-05-10 宁波舜宇智能科技有限公司 work piece detection method and device
CN106643483B (en) * 2016-09-28 2019-03-08 宁波舜宇智能科技有限公司 Workpiece inspection method and device
CN107514974A (en) * 2017-09-07 2017-12-26 唐冬香 A kind of method and system of lathe detection workpiece
CN107705293A (en) * 2017-09-14 2018-02-16 广东工业大学 A kind of hardware dimension measurement method based on CCD area array cameras vision-based detections
CN107767369A (en) * 2017-09-27 2018-03-06 杭州迈锐钶科技有限公司 A kind of the defects of buret detection method and device
CN108871185A (en) * 2018-05-10 2018-11-23 苏州大学 Method, apparatus, equipment and the computer readable storage medium of piece test
CN109211136A (en) * 2018-08-31 2019-01-15 广州大学 A kind of Watch glass cover board profile tolerance detection method
CN109990707A (en) * 2019-04-02 2019-07-09 天津工业大学 A kind of cut-parts eye knife detection method based on corner constraint
CN109990707B (en) * 2019-04-02 2021-04-02 天津工业大学 Corner point constraint-based cut-part eye knife detection method
CN111174703A (en) * 2020-02-24 2020-05-19 湖南工业大学 Non-contact size measurement method based on machine vision

Also Published As

Publication number Publication date
CN103499297B (en) 2016-01-13

Similar Documents

Publication Publication Date Title
CN103411553B (en) The quick calibrating method of multi-linear structured light vision sensors
Garcia et al. A combined temporal tracking and stereo-correlation technique for accurate measurement of 3D displacements: application to sheet metal forming
CN101520319B (en) Composite three-dimensional laser measurement system and measurement method
US10899014B2 (en) Multiple lens-based smart mechanical arm and positioning and assembly method thereof
CN101274432B (en) Apparatus for picking up objects
CN100561121C (en) Measurement method of planeness based on Flame Image Process and image recognition
Lahajnar et al. Machine vision system for inspecting electric plates
CN104034263B (en) A kind of non-contact measurement method of forging's block dimension
CN100458359C (en) Small-displacement measuring system in long-distance plane
CN104439698A (en) Calibration method and device used for laser processing system
CN100483116C (en) Method for detecting 3D defects on surface of belt material
CN100561120C (en) A kind of formation method of three-dimension measuring system
CN103471531B (en) The online non-contact measurement method of axial workpiece linearity
CN102853786B (en) Apparatus and method for detecting flatness
CN105783723B (en) Precision die surface processing accuracy detection device and method based on machine vision
CN100547351C (en) A kind of machine vision localization method
CN103712555B (en) Automotive frame pilot hole vision on-line measurement system and method thereof
CN104567727B (en) Global unified calibration method for linear structured light profile sensor through three-dimensional target
CN102435512B (en) Visual measurement method of plate forming limit curve (FLC)
CN103093223B (en) A kind of method for rapidly positioning of light spot image center
Vilaca et al. Calibration procedure for 3D measurement systems using two cameras and a laser line
CN106767524A (en) A kind of hydraulic spoon of blade detection method and device
CN104331896B (en) A kind of system calibrating method based on depth information
CN104897062B (en) A kind of vision measuring method and device of part antarafacial parallel hole shape position error
CN104657711B (en) A kind of readings of pointer type meters automatic identifying method of robust

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant