CN101936919B - Computer vision based glass quality detection device and measuring method thereof - Google Patents

Computer vision based glass quality detection device and measuring method thereof Download PDF

Info

Publication number
CN101936919B
CN101936919B CN2010102492479A CN201010249247A CN101936919B CN 101936919 B CN101936919 B CN 101936919B CN 2010102492479 A CN2010102492479 A CN 2010102492479A CN 201010249247 A CN201010249247 A CN 201010249247A CN 101936919 B CN101936919 B CN 101936919B
Authority
CN
China
Prior art keywords
black
glass
lattice
pixels point
corner pixels
Prior art date
Application number
CN2010102492479A
Other languages
Chinese (zh)
Other versions
CN101936919A (en
Inventor
陈�峰
王伟
冯定成
Original Assignee
清华大学
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 清华大学 filed Critical 清华大学
Priority to CN2010102492479A priority Critical patent/CN101936919B/en
Publication of CN101936919A publication Critical patent/CN101936919A/en
Application granted granted Critical
Publication of CN101936919B publication Critical patent/CN101936919B/en

Links

Abstract

The invention relates to a computer vision based glass quality detection device and a measuring method thereof. The device comprises a computer with a video capture card, a camera, a video cable connected with the computer and a template with black and while checks staggered longitudinally and horizontally. In the invention, the quality grade of glass is obtained through analyzing the distortion degree of a checker image generated by mapping a detected glass, i.e. the quality grade of the glass can be obtained by shooting the a black and white image which is reflected from the detected glass, transmitting the shot picture to the computer, analyzing a shot result in the computer, and assessing the quality grade of the glass according to a reflecting result. The invention can be used for automatically finishing a detection process of the glass quality without man power, thereby improving the automation level of glass production, and the glass detection process is accurate, quick and convenient; and moreover, the traditional computer vision technology is used as a tool, and the reflecting property of the glass is used as a quality assessment reference. The invention has the advantages of simple equipment, low cost, convenient mounting and easy operation.

Description

A kind of glass quality pick-up unit and measuring method thereof based on computer vision

Technical field

The invention belongs to glass quality detection technique field, be specifically related to a kind of glass quality pick-up unit and measuring method thereof based on computer vision.

Background technology

Be made up of zebra curtain, bazooka, interlock support and turning handle for the employed zebra method of the detection of glass quality detector at present, illuminating lamp tube throws light on the zebra curtain equably, and the observer is in the position apart from 9 meters on curtain; Pass through turning handle; The glass sample on the glass supporter is stood in rotation, sees through the distortion situation that glass sample is observed the zebra striped, measures the zebra-stripe distortion; This method need be put into glass the linkage support place of appointment when glass detects; And the deformation extent that needs a dead lift and manual observation zebra-stripe, the lack of uniform standard is not suitable in the large-scale glass production process.In the quality testing process, also need expend more human and material resources, automatization level is lower.

And the other a kind of glass quality detecting instrument that uses at present; It is made up of illuminator, worktable, detector, message handler four major parts of putting glass pieces to be measured: described illuminator by light source, beam treatment device and workpiece sensing before pretreater form; Described beam treatment device is meant the optical system that is complementary with described light source; Pretreater is made up of pressure gas clearer that has nozzle and the anti-reflection brush that the edge that is used for glass pieces to be measured strengthens light transmission before the described workpiece sensing; Above-mentioned each position component relation is: the light that described light source sends is through being radiated at behind the beam treatment device on the glass pieces to be measured that is positioned at described work piece platform; Described detector is gathered the image information of this glass pieces to be measured; Send described message handler to handle this information, storage and display image, and the shortcoming of this method is that required equipment is comparatively complicated, needs equipment such as special illuminator, light velocity processor, light-source system, pretreater; Equipment is installed complicated, and installation and maintenance is more loaded down with trivial details.

Summary of the invention

In order to overcome the deficiency that above-mentioned prior art exists; The object of the present invention is to provide a kind of glass quality pick-up unit and measuring method thereof based on computer vision; This device is by the computing machine that has video frequency collection card, camera and have vertically and the template of the staggered black and white grid of level is formed; Analysis through to the distortion degree of the gridiron pattern image of tested glass mapping obtains the glass quality grade, and promptly camera photographs the black white image that reflects in the tested glass, and picture shot will be sent in the computing machine; And in computing machine, shooting results is analyzed, assess the quality grade of glass according to reflection results; The present invention accomplishes the testing process of glass quality automatically like this, need not artificial the participation, has improved the automatization level of glass production; Make the glass testing process accurately, fast, convenient, and be instrument with existing computer vision technique, the reflective characteristic of utilizing glass is as the quality evaluation foundation; Equipment is simple; With low cost, easy for installation, easy operating.

In order to achieve the above object, the technical scheme that the present invention adopted is:

A kind of glass quality pick-up unit based on computer vision; Comprise and have the be interspersed black and white grid template 1 of black lattice and white square of vertical alternation sum level; Described black and white grid template 1 is placed horizontally at the tested glass 3 on the streamline lathe 2 towards correspondence; And this black and white grid template 1 surface tilts with respect to vertical direction, shape at a tilt angle, this angle of inclination can make described black and white grid template 1 be projected to tested glass 3 surfaces; Form black and white grid module view field; This view field is in the image pickup scope of the camera 4 of correspondence, and the input port of this video frequency collection card 5 of this camera 4 and the computing machine that has video frequency collection card 56 is connected, and Corner Detection module, black and white square detection module, glass detection module and quality assessment modules are arranged in the computing machine 6.

Described Corner Detection module is obtained the left first half brightness value A that each pixel in the black and white grid module map picture of video frequency collection card 5 input is divided equally 1,Right first half brightness value B 2,Left side Lower Half brightness value B 1And right Lower Half brightness value A 2, follow these four brightness values according to formula f=A 1+ A 2-B 1-B 2Obtain eigenwert f; Behind each pixel eigenwert f separately in having obtained black and white grid module map picture; According to preset angle point threshold value all eigenwert f are carried out binaryzation and non-maximum value inhibition processing then; Just can exclude the non-corner pixels point in the image, the net result that obtains is all black lattice and each the corner pixels point of white square in the black and white grid module map picture.

Described black and white square detection module is judged according to the eigenwert f of each corner pixels point; If the f value is a negative value, then its corresponding corner pixels point is category-A corner pixels point 7, if the f value be on the occasion of; Then its corresponding corner pixels point is category-B corner pixels point 8; Category-A corner pixels point 7 and category-B corner pixels point 8 transversely direction or longitudinal direction are staggered, and every group of four adjacent up and down corner pixels dot sequencies are coupled together, so every group of one four limit lattice of four angle points formations that are linked in sequence; Left upper end is that four limit lattice of category-A corner pixels point 7 are black four limit lattice, and left upper end is that four limit lattice of category-B corner pixels point 8 are white four limit lattice.

Described quality assessment modules is utilized method for marking connected region, and adjacent black four limit lattice and lattice zone, white four limit are coupled together, and constitutes continuous zone, and a corresponding tested glass 3 is represented in the continuum of an integral body.

Described glass detection module at first obtains the length of side L of the top margin of each the four limit lattice in the corresponding continuum of every block of tested glass 3 1, left side length of side L 2, right edge length of side L 3And the length of side L on base 4Size, and the upper left included angle A of these four limits lattice 1, upper right included angle A 2, the following included angle A in a left side 3And bottom right included angle A 4, utilize the dilatation degree α index computing formula of four limit lattice to calculate corresponding dilatation degree α again:

α = ( L 1 - L ‾ ) 2 + ( L 2 - L ‾ ) 2 + ( L 3 - L ‾ ) 2 + ( L 4 - L ‾ ) 2 4 / L ‾

Wherein The length of side L of expression top margin 1, left side length of side L 2, right edge length of side L 3And the length of side L on base 4Mean value;

And the torsional deformation degree β computing formula of four limit lattice:

β=(| A 1-90 ° |+| A 2-90 ° |+| A 3-90 ° |+| A 4-90 ° |)/360 ° of torsional deformation degree that calculate correspondence, the last dilatation degree α and the torsional deformation degree β that just can obtain all four limit lattice, the quality index of tested like this glass just is able to confirm.

The detection method of above-mentioned glass quality pick-up unit based on computer vision, step is following:

Step 1: have the be interspersed black and white grid template 1 of black lattice and white square of described vertical alternation sum level and be projected to tested glass 3 surfaces and form black and white grid module view fields; Start and have the computing machine 6 of video frequency collection card 5 and connect camera 4; Through camera 4 the black and white grid module view field that forms is sent to video frequency collection card 5, video frequency collection card 5 converts this black and white grid module view field into black and white grid module map and looks like to be transferred to the Corner Detection module in the computing machine 6 then;

Step 2: described Corner Detection module is at first obtained the left first half brightness value A that each pixel in the black and white grid module map picture is divided equally 1,Right first half brightness value B 2,Left side Lower Half brightness value B 1And right Lower Half brightness value A 2, follow these four brightness values according to formula f=A 1+ A 2-B 1-B 2Obtain eigenwert f; Behind each pixel eigenwert f separately in having obtained black and white grid module map picture; According to preset angle point threshold value all eigenwert f are carried out binaryzation and non-maximum value inhibition processing then; Just can exclude the non-corner pixels point in the image, the net result that obtains is all black lattice and each the corner pixels point of white square in the black and white grid module map picture;

Step 3: through the black and white square detection module in the computing machine 6; Eigenwert f according to each corner pixels point judges; If the f value is a negative value, then its corresponding corner pixels point is category-A corner pixels point 7, if the f value be on the occasion of; Then its corresponding corner pixels point is category-B corner pixels point 8; Category-A corner pixels point 7 and category-B corner pixels point 8 transversely direction or longitudinal direction are staggered, and every group of four adjacent up and down corner pixels dot sequencies are coupled together, so every group of one four limit lattice of four angle points formations that are linked in sequence; Left upper end is that four limit lattice of category-A corner pixels point 7 are black four limit lattice, and left upper end is that four limit lattice of category-B corner pixels point 8 are white four limit lattice;

Step 4: through the quality assessment modules in the computing machine 6, utilize method for marking connected region, adjacent black four limit lattice and lattice zone, white four limit are coupled together, constitute continuous zone, a corresponding tested glass 3 is represented in the continuum of an integral body;

Step 5:, at first obtain the length of side L of the top margin of each the four limit lattice in the corresponding continuum of every block of tested glass 3 through the glass detection module in the computing machine 6 1, left side length of side L 2, right edge length of side L 3And the length of side L on base 4Size, and the upper left included angle A of these four limits lattice 1, upper right included angle A 2, the following included angle A in a left side 3And bottom right included angle A 4, utilize the dilatation degree α index computing formula of four limit lattice to calculate the dilatation degree α of four corresponding limit lattice again:

α = ( L 1 - L ‾ ) 2 + ( L 2 - L ‾ ) 2 + ( L 3 - L ‾ ) 2 + ( L 4 - L ‾ ) 2 4 / L ‾

Wherein The length of side L of expression top margin 1, left side length of side L 2, right edge length of side L 3And the length of side L on base 4Mean value;

And the torsional deformation degree β computing formula of four limit lattice:

β=(| A 1-90 ° |+| A 2-90 ° |+| A 3-90 ° |+| A 4-90 ° |)/360 ° of torsional deformation degree that calculate correspondence, the last dilatation degree α and the torsional deformation degree β that just can obtain all four limit lattice, the quality index of tested like this glass just is able to confirm.

A kind of glass quality pick-up unit and measuring method thereof of the present invention based on computer vision; This device is by the computing machine that has video frequency collection card 6, camera 4 and have vertically and the template 1 of the staggered black and white grid of level is formed; Analysis through to the distortion degree of the gridiron pattern image of tested glass 3 mappings obtains the glass quality grade; Be that camera photographs the black white image that reflects in the tested glass 3; Picture shot will be sent in the computing machine 6, and in computing machine 6, shooting results is analyzed, and assesses the quality grade of glass according to reflection results; The present invention accomplishes the testing process of glass quality automatically like this, need not artificial the participation, has improved the automatization level of glass production; Make the glass testing process accurately, fast, convenient, and be instrument with existing computer vision technique, the reflective characteristic of utilizing glass is as the quality evaluation foundation; Equipment is simple; With low cost, easy for installation, easy operating.

Description of drawings

Fig. 1 is the work synoptic diagram of the glass quality pick-up unit based on computer vision of the present invention, and the dotted line that wherein is connected with black and white grid template is the projecting direction of black and white grid template, and the dotted line that wherein is connected with camera is the direction of camera shooting.

Fig. 2 is a category-A corner pixels point of the present invention.

Fig. 3 is a category-B corner pixels point of the present invention.

Embodiment

Below in conjunction with accompanying drawing the present invention is done more detailed explanation.

As shown in Figure 1; Glass quality pick-up unit based on computer vision; Comprise having the be interspersed black and white grid template 1 of black lattice and white square of vertical alternation sum level, described black and white grid template 1 is placed horizontally at the tested glass 3 on the streamline lathe 2 towards correspondence, and this black and white grid template 1 surface tilts with respect to vertical direction; Shape at a tilt angle; This angle of inclination can make described black and white grid template 1 be projected to tested glass 3 surfaces, forms black and white grid module view field, and this view field is in the image pickup scope of the camera 4 of correspondence; The input port of this video frequency collection card 5 of this camera 4 and the computing machine that has video frequency collection card 56 is connected, and Corner Detection module, black and white square detection module, glass detection module and quality assessment modules are arranged in the computing machine 6.

The detection method based on the glass quality pick-up unit of computer vision of present embodiment, step is following:

Step 1: have the be interspersed black and white grid template 1 of black lattice and white square of described vertical alternation sum level and be projected to tested glass 3 surfaces and form black and white grid module view fields; Start and have the computing machine 6 of video frequency collection card 5 and connect camera 4; Through camera 4 the black and white grid module view field that forms is sent to video frequency collection card 5, video frequency collection card 5 converts this black and white grid module view field into black and white grid module map and looks like to be transferred to the Corner Detection module in the computing machine 6 then;

Step 2: described Corner Detection module is at first obtained the left first half brightness value A that each pixel in the black and white grid module map picture is divided equally 1,Right first half brightness value B 2,Left side Lower Half brightness value B 1And right Lower Half brightness value A 2, follow these four brightness values according to formula f=A 1+ A 2-B 1-B 2Obtain eigenwert f; Behind each pixel eigenwert f separately in having obtained black and white grid module map picture; According to preset angle point threshold value all eigenwert f are carried out binaryzation and non-maximum value inhibition processing then; Just can exclude the non-corner pixels point in the image, the net result that obtains is all black lattice and each the corner pixels point of white square in the black and white grid module map picture;

Step 3: through the black and white square detection module in the computing machine 6; Eigenwert f according to each corner pixels point judges; If the f value is a negative value, then its corresponding corner pixels point is category-A corner pixels point 7, if the f value be on the occasion of; Then its corresponding corner pixels point is category-B corner pixels point 8; Category-A corner pixels point 7 and category-B corner pixels point 8 transversely direction or longitudinal direction are staggered, and every group of four adjacent up and down corner pixels dot sequencies are coupled together, so every group of one four limit lattice of four angle points formations that are linked in sequence; Left upper end is that four limit lattice of category-A corner pixels point 7 are black four limit lattice, and left upper end is that four limit lattice of category-B corner pixels point 8 are white four limit lattice;

Step 4: through the quality assessment modules in the computing machine 6, utilize method for marking connected region, adjacent black four limit lattice and lattice zone, white four limit are coupled together, constitute continuous zone, a corresponding tested glass 3 is represented in the continuum of an integral body;

Step 5:, at first obtain the length of side L of the top margin of each the four limit lattice in the corresponding continuum of every block of tested glass 3 through the glass detection module in the computing machine 6 1, left side length of side L 2, right edge length of side L 3And the length of side L on base 4Size, and the upper left included angle A of these four limits lattice 1, upper right included angle A 2, the following included angle A in a left side 3And bottom right included angle A 4, utilize the dilatation degree α index computing formula of four limit lattice to calculate corresponding dilatation degree α again:

α = ( L 1 - L ‾ ) 2 + ( L 2 - L ‾ ) 2 + ( L 3 - L ‾ ) 2 + ( L 4 - L ‾ ) 2 4 / L ‾

Wherein The length of side L of expression top margin 1, left side length of side L 2, right edge length of side L 3And the length of side L on base 4Mean value;

And the torsional deformation degree β computing formula of four limit lattice:

β=(| A 1-90 ° |+| A 2-90 ° |+| A 3-90 ° |+| A 4-90 ° |)/360 ° of torsional deformation degree that calculate correspondence, the last dilatation degree α and the torsional deformation degree β that just can obtain all four limit lattice, the quality index of tested like this glass just is able to confirm.

A kind of glass quality pick-up unit and measuring method thereof of the present invention based on computer vision; This device is by the computing machine that has video frequency collection card 6, camera 4 and have vertically and the template 1 of the staggered black and white grid of level is formed; Analysis through to the distortion degree of the gridiron pattern image of tested glass 3 mappings obtains the glass quality grade; Be that camera photographs the black white image that reflects in the tested glass 3; Picture shot will be sent in the computing machine 6, and in computing machine 6, shooting results is analyzed, and assesses the quality grade of glass according to reflection results; The present invention accomplishes the testing process of glass quality automatically like this, need not artificial the participation, has improved the automatization level of glass production; Make the glass testing process accurately, fast, convenient, and be instrument with existing computer vision technique, the reflective characteristic of utilizing glass is as the quality evaluation foundation; Equipment is simple; With low cost, easy for installation, easy operating.

Claims (2)

1. glass quality pick-up unit based on computer vision; It is characterized in that: comprise having the be interspersed black and white grid template (1) of black lattice and white square of vertical alternation sum level; Described black and white grid template (1) is placed horizontally at the tested glass (3) on the streamline lathe (2) towards correspondence; And this black and white grid template (1) surface tilts with respect to vertical direction; Shape at a tilt angle; This angle of inclination can make described black and white grid template (1) be projected to tested glass (3) surface; Form black and white grid module view field, this view field is in the image pickup scope of the camera (4) of correspondence, and the input port of this video frequency collection card (5) of this camera (4) and the computing machine that has video frequency collection card (5) (6) is connected; Corner Detection module, black and white square detection module, glass detection module and quality assessment modules are arranged in the computing machine (6), and described Corner Detection module is obtained the left first half brightness value A that each pixel in the black and white grid module map picture of video frequency collection card (5) input is divided equally 1,Right first half brightness value B 2,Left side Lower Half brightness value B 1And right Lower Half brightness value A 2, follow these four brightness values according to formula f=A 1+ A 2-B 1-B 2Obtain eigenwert f, behind each pixel eigenwert f separately in having obtained black and white grid module map picture, according to preset angle point threshold value all eigenwert f are carried out binaryzation and non-maximum value inhibition processing then; Just can exclude the non-corner pixels point in the image, the net result that obtains is all black lattice and each the corner pixels point of white square in the black and white grid module map picture, and described black and white square detection module is judged according to the eigenwert f of each corner pixels point; If the f value is a negative value; Then its corresponding corner pixels point is a category-A corner pixels point (7), if the f value be on the occasion of, then its corresponding corner pixels point is a category-B corner pixels point (8); Category-A corner pixels point (7) and category-B corner pixels point (8) transversely direction or longitudinal direction are staggered; Every group of four adjacent up and down corner pixels dot sequencies are coupled together, and every group of four angle points that are linked in sequence constitute one four limit lattice like this, and left upper end is that four limit lattice of category-A corner pixels point (7) are black four limit lattice; Left upper end is that four limit lattice of category-B corner pixels point (8) are white four limit lattice; Described quality assessment modules is utilized method for marking connected region, and adjacent black four limit lattice and lattice zone, white four limit are coupled together; Constitute continuous zone; A corresponding tested glass (3) is represented in the continuum of an integral body, and described glass detection module at first obtains the length of side L of the top margin of each the four limit lattice in the corresponding continuum of every block of tested glass (3) 1, left side length of side L 2, right edge length of side L 3And the length of side L on base 4Size, and the upper left included angle A of these four limits lattice 1, upper right included angle A 2, the following included angle A in a left side 3And bottom right included angle A 4, utilize the dilatation degree α index computing formula of four limit lattice to calculate corresponding dilatation degree α again:
Wherein The length of side L of expression top margin 1, left side length of side L 2, right edge length of side L 3And the length of side L on base 4Mean value;
And the torsional deformation degree β computing formula of four limit lattice:
β=(| A 1-90 ° |+| A 2-90 ° |+| A 3-90 ° |+| A 4-90 ° |)/360 ° of torsional deformation degree that calculate correspondence, the last dilatation degree α and the torsional deformation degree β that just can obtain all four limit lattice, the quality index of tested like this glass just is able to confirm.
2. the measuring method of the glass quality pick-up unit based on computer vision according to claim 1 is characterized in that step is following:
Step 1: have the be interspersed black and white grid template (1) of black lattice and white square of described vertical alternation sum level and be projected to tested glass (3) surface and form black and white grid module view field; Start and have the computing machine (6) of video frequency collection card (5) and connect camera (4); Through camera (4) the black and white grid module view field that forms is sent to video frequency collection card (5), video frequency collection card (5) converts this black and white grid module view field into black and white grid module map and looks like to be transferred to the Corner Detection module in the computing machine (6) then;
Step 2: described Corner Detection module is at first obtained the left first half brightness value A that each pixel in the black and white grid module map picture is divided equally 1,Right first half brightness value B 2,Left side Lower Half brightness value B 1And right Lower Half brightness value A 2, follow these four brightness values according to formula f=A 1+ A 2-B 1-B 2Obtain eigenwert f; Behind each pixel eigenwert f separately in having obtained black and white grid module map picture; According to preset angle point threshold value all eigenwert f are carried out binaryzation and non-maximum value inhibition processing then; Just can exclude the non-corner pixels point in the image, the net result that obtains is all black lattice and each the corner pixels point of white square in the black and white grid module map picture;
Step 3: through the black and white square detection module in the computing machine (6); Eigenwert f according to each corner pixels point judges; If the f value is a negative value, then its corresponding corner pixels point is a category-A corner pixels point (7), if the f value be on the occasion of; Then its corresponding corner pixels point is a category-B corner pixels point (8); Category-A corner pixels point (7) and category-B corner pixels point (8) transversely direction or longitudinal direction are staggered, and every group of four adjacent up and down corner pixels dot sequencies are coupled together, and every group of four angle points that are linked in sequence constitute one four limit lattice like this; Left upper end is that four limit lattice of category-A corner pixels point (7) are black four limit lattice, and left upper end is that four limit lattice of category-B corner pixels point (8) are white four limit lattice;
Step 4: through the quality assessment modules in the computing machine (6); Utilize method for marking connected region; Adjacent black four limit lattice and lattice zone, white four limit are coupled together, constitute continuous zone, a corresponding tested glass (3) is represented in the continuum of an integral body;
Step 5:, at first obtain the length of side L of the top margin of each the four limit lattice in the corresponding continuum of every block of tested glass (3) through the glass detection module in the computing machine (6) 1, left side length of side L 2, right edge length of side L 3And the length of side L on base 4Size, and the upper left included angle A of these four limits lattice 1, upper right included angle A 2, the following included angle A in a left side 3And bottom right included angle A 4, utilize the dilatation degree α index computing formula of four limit lattice to calculate corresponding dilatation degree α again:
Wherein The length of side L of expression top margin 1, left side length of side L 2, right edge length of side L 3And the length of side L on base 4Mean value;
And the torsional deformation degree β computing formula of four limit lattice:
β=(| A 1-90 ° |+| A 2-90 ° |+| A 3-90 ° |+| A 4-90 ° |)/360 ° of torsional deformation degree that calculate correspondence, the last dilatation degree α and the torsional deformation degree β that just can obtain all four limit lattice, the quality index of tested like this glass just is able to confirm.
CN2010102492479A 2010-08-06 2010-08-06 Computer vision based glass quality detection device and measuring method thereof CN101936919B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN2010102492479A CN101936919B (en) 2010-08-06 2010-08-06 Computer vision based glass quality detection device and measuring method thereof

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN2010102492479A CN101936919B (en) 2010-08-06 2010-08-06 Computer vision based glass quality detection device and measuring method thereof

Publications (2)

Publication Number Publication Date
CN101936919A CN101936919A (en) 2011-01-05
CN101936919B true CN101936919B (en) 2012-02-22

Family

ID=43390361

Family Applications (1)

Application Number Title Priority Date Filing Date
CN2010102492479A CN101936919B (en) 2010-08-06 2010-08-06 Computer vision based glass quality detection device and measuring method thereof

Country Status (1)

Country Link
CN (1) CN101936919B (en)

Families Citing this family (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
FR2974414B1 (en) * 2011-04-22 2013-04-12 Saint Gobain METHOD FOR ANALYZING THE QUALITY OF A GLAZING
US9488597B2 (en) * 2011-11-30 2016-11-08 Corning Incorporated Apparatus and methods for determining surface compliance for a glass surface
CN103925886A (en) * 2013-01-15 2014-07-16 中芯国际集成电路制造(上海)有限公司 Wafer deformation detection system and method
CN104316543B (en) * 2014-10-10 2016-10-05 南京大树智能科技股份有限公司 A kind of tobacco bale cellophane detection method
CN106091997B (en) * 2016-06-01 2019-12-17 广州视睿电子科技有限公司 Method, device and system for testing glass performance
CN106295713A (en) * 2016-08-19 2017-01-04 华南理工大学 The article condition label recognition system of a kind of computer-oriented visual cognition and method
CN106404803A (en) * 2016-09-09 2017-02-15 蚌埠中建材信息显示材料有限公司 Detection method for appearance quality of ultrathin float glass
CN107203994A (en) * 2017-06-08 2017-09-26 广东嘉铭智能科技有限公司 A kind of method and apparatus of globoidal glass curvature measuring

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5003615A (en) * 1988-12-01 1991-03-26 Harris Semiconductor Patents, Inc. Optoelectronic system for determining surface irregularities of a workpiece having a nominally plane reflective surface
CN101660968A (en) * 2009-09-25 2010-03-03 中国科学院上海光学精密机械研究所 Device and method for detecting optical performance of transparent piece

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP3688520B2 (en) * 1999-07-06 2005-08-31 住友化学株式会社 Surface inspection apparatus and surface inspection method
JP3899915B2 (en) * 2001-12-06 2007-03-28 旭硝子株式会社 Optical strain evaluation method and evaluation apparatus
JP2004333230A (en) * 2003-05-02 2004-11-25 Fuji Photo Film Co Ltd Method for measuring surface shape of medium and information recording medium

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5003615A (en) * 1988-12-01 1991-03-26 Harris Semiconductor Patents, Inc. Optoelectronic system for determining surface irregularities of a workpiece having a nominally plane reflective surface
CN101660968A (en) * 2009-09-25 2010-03-03 中国科学院上海光学精密机械研究所 Device and method for detecting optical performance of transparent piece

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
JP特开2001-21332A 2001.01.26
JP特开2003-172713A 2003.06.20
JP特开2004-333230A 2004.11.25

Also Published As

Publication number Publication date
CN101936919A (en) 2011-01-05

Similar Documents

Publication Publication Date Title
CN103759638B (en) A kind of part detection method
CA2467624C (en) Procedure and appliance for touch-free examinations of objects, especially regarding their surface character
US9279662B2 (en) Laser scanner
CN1156758C (en) Method and device for calibrating project image from computer
EP2102587B1 (en) Method of automated quantitative analysis of distortion shaped vehicle glass by reflected optical imaging
KR101326991B1 (en) Apparatus and method for sensing property of road surface
CN106017325B (en) A kind of noncontact optical measurement method of improved complex surface and irregularly shaped object volume
CN102713582B (en) Inclusion detection in polished gemstones
KR101830679B1 (en) Apparatus for testing a display panel and method thereof
US8081840B2 (en) Appliance for controlling transparent or reflective elements
US8736688B2 (en) Method and device for analyzing the optical quality of a transparent substrate
WO2004070693A3 (en) Method and apparatus for optical inspection of a display
US9454024B1 (en) Composite repairing apparatus for liquid crystal panel
CN103499297A (en) CCD (Charge Coupled Device)-based high-accuracy measuring method
CN104554341B (en) The system and method for detecting rail smooth degree
CN104101603B (en) Steel strip surface defect detection method
CN103065520B (en) Detection system for backing car into storage and detection method thereof
JP5010001B2 (en) 3D image display panel evaluation method and system
CN102565072B (en) Stereoscopic vision on-line detection method for surface crack of stretched aluminum alloy plate
CN102171554A (en) Device for analysing the surface of a substrate
CN1804927A (en) Omnibearing visual sensor based road monitoring apparatus
CN101620816B (en) Method and system for evaluating stereoscopic image display panel
CN102692188B (en) Dynamic crack length measurement method for machine vision fatigue crack propagation test
CN102636313A (en) Leakage source detecting device based on infrared thermal imaging processing
CN1460967A (en) Image quality analysis method and system for display device

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
EXPY Termination of patent right or utility model
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20120222

Termination date: 20140806