CN106767564A - A kind of detection method for being applied to phone housing surface roughness - Google Patents
A kind of detection method for being applied to phone housing surface roughness Download PDFInfo
- Publication number
- CN106767564A CN106767564A CN201610953601.3A CN201610953601A CN106767564A CN 106767564 A CN106767564 A CN 106767564A CN 201610953601 A CN201610953601 A CN 201610953601A CN 106767564 A CN106767564 A CN 106767564A
- Authority
- CN
- China
- Prior art keywords
- image
- phone housing
- surface roughness
- detection
- housing surface
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B11/00—Measuring arrangements characterised by the use of optical techniques
- G01B11/30—Measuring arrangements characterised by the use of optical techniques for measuring roughness or irregularity of surfaces
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B11/00—Measuring arrangements characterised by the use of optical techniques
- G01B11/30—Measuring arrangements characterised by the use of optical techniques for measuring roughness or irregularity of surfaces
- G01B11/303—Measuring arrangements characterised by the use of optical techniques for measuring roughness or irregularity of surfaces using photoelectric detection means
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Length Measuring Devices By Optical Means (AREA)
- Image Analysis (AREA)
Abstract
The present invention relates to a kind of detection method for being applied to phone housing surface roughness, for phone housing, the detection of different phase has different roughness and in real time on-line monitoring in automatically grinding polishing process, the algorithm divides to mobile phone housing surface area sample first, then the image after sampling is zoomed in and out, gray processing, binaryzation etc. is pre-processed, the interference of noise on image is reduced with medium filtering, rim detection is carried out finally according to a kind of auto-adaptable image edge detection Canny methods, extract the pixel line-spacing mean difference of crest and trough, carry out curve fitting and modified weight, obtain an accurate roughness grade.
Description
Technical field
The present invention relates to the image processing field of computer vision, more particularly to one kind is applied to phone housing rough surface
The detection method of degree.
Background technology
This part of, recruitment that the inventive method is related to the image procossing of computer vision, particularly image basic handling
Industry CCD camera detects the surface roughness curve obtained by light-cutting microscope, so as to obtaining roughness grade.
The method most close with the present invention has Yao Songli [1] et al. to be proposed for the roughness measurement on phone housing surface
A kind of Surface roughness measurement system based on machine vision, cuts from light with image processing techniques and surface is extracted in micro-image
Contour signal calculates roughness evaluation parameter.
Citation:
[1] such as Yao Songli, department's sword merit, Ren Hongli is based on Surface roughness measurement system design [J] works of machine vision
Industry control computer .2015,28 (6):71-72.
[2] Liu Bin, Feng Qibo, rectify extraction side Survey of measurement methods for surface roughness [J] optical instruments, 2004,26 (5):
54-55.
[3] Yuan Hui is beautiful waits Noncontact Surface Roughenss Measuring Instrument [J] Haerbin Scientific and Technological Univ. journal, 1995,19
(6):30-34.
The content of the invention
Current existing roughness technology is divided into contact type measurement and non-contact measurement, and contact type measurement is mainly contact pilotage
Method, directly contact measured surface measures more stable, and contact type measurement has very big defect, is in particular in:1. to high-precision
There is scuffing destruction on degree surface and soft metal surface;2. limited by stylus tip arc radius, its certainty of measurement is limited;
3. because of contact pilotage abrasion and the limitation of measuring speed, it is impossible to realize On-line sampling system.For this problem, the present invention is directed to mobile phone
The detection of case surface roughness, the light-cutting microscope first by CCD industrial cameras and noncontact up-to-date style extracts workpiece
Surface image, then zooms in and out, the rim detection of gray processing, binaryzation, medium filtering, self adaptation etc. is to surface image
The image processing work of row, realizes the On-line sampling system of automatic roughness measurement, solves this problem.
The present invention proposes a kind of detection method for being applied to phone housing surface roughness, first to mobile phone case surface area
Domain divides sampling, and then the image after sampling is pre-processed, and the interference of noise on image, last root are reduced with medium filtering
Rim detection is carried out according to auto-adaptable image edge detection Canny methods, the pixel line-spacing mean difference of crest and trough is extracted, entered
Row curve matching and modified weight, obtain roughness grade.
Further, it is described pretreatment include zoom in and out, gray processing, binaryzation the step of.
Further, the auto-adaptable image edge detection Canny methods are comprised the following steps:
Use Gaussian filter smoothed image;
Amplitude and the direction of gradient are calculated with the finite difference of single order local derviation;
To gradient magnitude application non-maxima suppression;
Detected with double threshold algorithms and link edge.
Further, the hardware device of collection image is light-cutting microscope and CCD industrial cameras.
Further, phone housing range of surface roughness is set to 6.3-0.8.
Brief description of the drawings
Fig. 1 is roughness detecting method FB(flow block);
Fig. 2A is mobile phone sample zoning figure;
Fig. 2 B are specimen sample result curve figure that roughness is 6.3,3.2,1.6,0.8;
Fig. 3 is gray processing result figure;
Fig. 4 is binaryzation result figure;
Fig. 5 is figure after medium filtering;
Fig. 6 is edge detection results figure;
Fig. 7 A are 6.3 roughness measurement result figures;
Fig. 7 B are 3.2 roughness measurement result figures;
Fig. 7 C are 1.6 roughness measurement result figures;
Fig. 7 D are 0.8 roughness measurement result figure.
Specific embodiment
Below in conjunction with accompanying drawing, the present invention is described in detail:
A kind of detection method for being applied to phone housing surface roughness, divides to mobile phone housing surface area adopt first
Sample, then pre-processes to the image after sampling, the interference of noise on image is reduced with medium filtering, finally according to self adaptation
Rim detection Canny methods carry out rim detection, extract the pixel line-spacing mean difference of crest and trough, carry out curve plan
Close and modified weight, obtain roughness grade.Fig. 1 is phone housing roughness detecting method FB(flow block).Mainly include
Seven aspects such as zoning sampling, image preprocessing, gray processing, binaryzation, medium filtering.
Area sampling:Because the hardware device for gathering image is light-cutting microscope and CCD industrial cameras, in order to measure
Surface roughness result it is more accurate, carry out carrying out multiple repairing weld on phone housing surface, for the image of multiple repairing weld
Subsequent treatment is carried out, the error of experiment is reduced.Behind zoning as shown in Figure 2 A.Here with 14 times of object lens, roughness be 6.3,
3.2nd, as a example by 1.6,0.8, the picture after being sampled is as shown in Figure 2 B.
Image preprocessing:Light-cutting microscope has four object lens, is respectively 7 times, 14 times, 30 times, 60 times.Machine is cut in light
The picture format clapped in CCD picture shooting softwares is 2048*1536, because picture is than larger, during process test,
With the specification that picture is reduced in opencv, the unified normal pictures for being contracted to 800*600.Image scaling when firstly the need of meter
The size of image after scaling is calculated, if newWidth, newHeight are the wide and height of the image after scaling, width, height are
The width and height of original image, K so have for scaling:
NewWidth=K*width
NewHeight=K*height (1)
Image gray processing:Coloured image includes substantial amounts of colouring information, to storage and the disposal ability requirement ratio of system
It is higher, and it is unfavorable for the calculating of surface roughness.And just it is called the ash of image by the process that coloured image is converted to gray level image
Degreeization treatment.Gray-scale map is the image without color information containing only monochrome information, although eliminate the information of coloured image, but
Otherwise information is but amplified, and image intensity value is represented with such as formula (2):
F=0.3R+0.59G+0.11B (2)
Wherein R, G, B are respectively the how corresponding colour information value of position pixel, and F is the gray scale of picture correspondence position pixel
Value.Switch to gray level image by by coloured image, so as to improve the contrast of entire image, some figures not observed originally
As details may be highlighted.Image is as shown in Figure 3 after gray processing.
Image binaryzation:Correspondence Pyatyi gray-scale map, binary conversion treatment is a kind of gray proces algorithm, for the threshold for giving
Gray scale is become white point by value, program more than the point of given threshold value, and point in addition is changed into stain, and image is changed into only black after treatment
The bianry image of Bai Erse.It is defined as follows for binaryzation:
But although due to the difference of light, distribution of color is not necessarily identical, substantially our sensitizing range
Then think that the pixel is set to white after binaryzation.Binary image after so processing is as shown in Figure 4.
Medium filtering:Medium filtering SM (Standard Median Filter) is a kind of with the non-of less edge blurry
Linear filter method, can not only remove or reduce random noise and impulse disturbances, moreover it is possible to preferably retain image border letter
Breath.This algorithm depends on quick sorting algorithm, and its basic thought is any selection one in the element set to be sorted
Simultaneously with other elements be compared for it by individual element, before all elements smaller than this element all are placed on into it, by all ratios
Its big element is put after it;After by a minor sort, can demarcate as the position where the element, set is divided into 2
Part;Then repeat said process to remaining 2 parts to be ranked up, untill each section only remains next element;
After the completion of all sequences, the value (i.e. so-called intermediate value) of centrally located element in the set after sequence is taken as output
Value.Traditional medium filtering can be defined as:
G (x, y)=med { f (xi, yj) (i, j) ∈ M (4)
Wherein g (x, y) is exported for medium filtering, f (xi, yj) it is the pixel (x of imagei, yj) gray value, M is template window
Mouthful.First can the information of defect is more obvious with an expansive working before medium filtering, as shown in Figure 5.
Rim detection:Canny edge detection operators its essence is smoothing operation is made with 1 quasi-Gaussian function, then with band
The first order differential operator positioning derivative maximum in direction, obtains Gaussian template derivative approximation, in theory very according to variational method
Close to the best edge operator that 4 linear combination of exponential functions are formed, smoothing processing, therefore tool are made to image using Gaussian function
There is stronger noise removal capability.Canny operators are concretely comprised the following steps
(1) Gaussian filter smoothed image is used;
(2) amplitude and the direction of gradient are calculated with the finite difference of single order local derviation;
(3) to gradient magnitude application non-maxima suppression;
(4) detected and link edge with double threshold algorithms.
According to the result after above-mentioned medium filtering, the edge detection results after being detected with Canny operators are as shown in Figure 6.
Curve matching and modified weight:By after a series of filtering and morphology operations, we have extracted finally
Curve map.Because curve map is obtained by ccd video camera sampling, the inclined phenomenon of curve is had unavoidably, can so cause
Than larger error.Least square fitting curve is first used herein, data is then carried out according to the pseudocurve for obtaining and is repaiied
Just so that curve map is in the same horizontal line, it is ensured that precision, the error of experiment is reduced.Can be easily using least square method
Try to achieve unknown data, and cause that the quadratic sum of error between these data tried to achieve and real data is minimum.Least square
The formula of method is as follows:
After being modified using least square fitting curve, we will extract 5 crest values and 5 ripples to curve map
Valley, so as to obtain a phone housing mean height of surface Rz.It is last to be measured due to having some unavoidably in measurement process
Error, so finally to carry out a slight modified weight to last result, reduce the error that detection process is caused.By
Then sampled some pictures, for a data group, we can obtain the data more than comparing, from outside multiple mobile phones
Shell mean height of surface Rz first removes several extremely high value and several extremely low values, sets a weights for self adaptation last to revise
Result parameter, reduces detection error.
It is applied in the detection of phone housing surface roughness, for best technology, it is impossible to reach online reality
When monitor, often to waste substantial amounts of man power and material and go to realize manual detection so have great inconvenience to automatic detection.
And the present invention is in original technical foundation, mobile phone case surface is carried out first region sampling several, after preservation
Image zooms in and out reduction amount of calculation, and medium filtering and rim detection are added on the basis of the image procossing of gray processing and binaryzation,
Reduce CCD camera and extract the noise jamming caused during image, finally by curve matching and deviation weight amendment, be relatively defined
The numerical value of true surface roughness.
There are several processes to cause different roughness during automatically grinding is polished according to phone housing, for this
Several situations, four kinds of grades of roughness are divided into by various process phone housing, are respectively 6.3,3.2,1.6,0.8 roughness
Detection case is respectively as shown in accompanying drawing 7A-7D.
Above-described embodiment is the present invention preferably implementation method, but embodiments of the present invention are not by above-described embodiment
Limitation, it is other it is any without departing from Spirit Essence of the invention and the change, modification, replacement made under principle, combine, simplification,
Equivalent substitute mode is should be, is included within protection scope of the present invention.
Claims (5)
1. a kind of detection method for being applied to phone housing surface roughness, it is characterised in that:First to mobile phone case surface area
Domain divides sampling, and then the image after sampling is pre-processed, and the interference of noise on image, last root are reduced with medium filtering
Rim detection is carried out according to auto-adaptable image edge detection Canny methods, the pixel line-spacing mean difference of crest and trough is extracted, entered
Row curve matching and modified weight, obtain roughness grade.
2. the detection method of phone housing surface roughness is applied to as claimed in claim 1, it is characterised in that the pre- place
Reason include zoom in and out, gray processing, binaryzation the step of.
3. the detection method of phone housing surface roughness is applied to as claimed in claim 1, it is characterised in that described adaptive
Rim detection Canny methods are answered to comprise the following steps:
Use Gaussian filter smoothed image;
Amplitude and the direction of gradient are calculated with the finite difference of single order local derviation;
To gradient magnitude application non-maxima suppression;
Detected with double threshold algorithms and link edge.
4. the detection method of phone housing surface roughness is applied to as claimed in claim 1, it is characterised in that collection image
Hardware device be light-cutting microscope and CCD industrial cameras.
5. the detection method of phone housing surface roughness is applied to as claimed in claim 1, it is characterised in that phone housing
Range of surface roughness is set to 6.3-0.8.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610953601.3A CN106767564A (en) | 2016-11-03 | 2016-11-03 | A kind of detection method for being applied to phone housing surface roughness |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610953601.3A CN106767564A (en) | 2016-11-03 | 2016-11-03 | A kind of detection method for being applied to phone housing surface roughness |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106767564A true CN106767564A (en) | 2017-05-31 |
Family
ID=58972657
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610953601.3A Pending CN106767564A (en) | 2016-11-03 | 2016-11-03 | A kind of detection method for being applied to phone housing surface roughness |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106767564A (en) |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109190642A (en) * | 2018-09-04 | 2019-01-11 | 华中科技大学 | The method for extracting surface characteristics using high-order Gauss regression filtering and Radon transformation |
CN109357636A (en) * | 2018-12-10 | 2019-02-19 | 电子科技大学 | A kind of phase amplification calculation method based on black part structure light scan |
CN109408888A (en) * | 2018-11-27 | 2019-03-01 | 广东工业大学 | A kind of roughness calculation method, computer readable storage medium and the terminal on two-dimensional cutting surface |
CN110211100A (en) * | 2019-05-20 | 2019-09-06 | 浙江大学 | A kind of foot measurement method of parameters based on image |
CN112050756A (en) * | 2020-09-04 | 2020-12-08 | 南通大学 | Rock ore slice and resin target surface flatness recognition processing method |
CN113701632A (en) * | 2021-09-01 | 2021-11-26 | 威海北洋电气集团股份有限公司 | Thread detection method based on difference value |
CN114046749A (en) * | 2021-10-26 | 2022-02-15 | 刘红霞 | Method and system for detecting roughness of point-shaped pit joint surface of precast concrete member |
CN114331923A (en) * | 2022-03-11 | 2022-04-12 | 中国空气动力研究与发展中心低速空气动力研究所 | Improved Canny algorithm-based bubble contour extraction method in ice structure |
CN114677340A (en) * | 2022-03-14 | 2022-06-28 | 上海第二工业大学 | Concrete surface roughness detection method based on image edge |
CN117523502A (en) * | 2024-01-08 | 2024-02-06 | 东莞市杰瑞智能科技有限公司 | Urban road rubbish intelligent monitoring system based on machine vision |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101804720A (en) * | 2010-03-16 | 2010-08-18 | 浙江博玛数码电子有限公司 | Vision-based online digital electronic engraving quality monitoring method and device |
CN101968347A (en) * | 2010-08-31 | 2011-02-09 | 苏州大学 | Method for extracting surface flatness of flat granular objects |
CN102564593A (en) * | 2011-12-30 | 2012-07-11 | 河海大学常州校区 | Plant growth condition monitoring system based on compute vision and internet of things |
WO2014134993A1 (en) * | 2013-03-04 | 2014-09-12 | 北京联想软件有限公司 | Target object information acquisition method and electronic device |
CN105526887A (en) * | 2016-02-23 | 2016-04-27 | 东华大学 | Non-contact type fabric surface roughness test method and test device |
-
2016
- 2016-11-03 CN CN201610953601.3A patent/CN106767564A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101804720A (en) * | 2010-03-16 | 2010-08-18 | 浙江博玛数码电子有限公司 | Vision-based online digital electronic engraving quality monitoring method and device |
CN101968347A (en) * | 2010-08-31 | 2011-02-09 | 苏州大学 | Method for extracting surface flatness of flat granular objects |
CN102564593A (en) * | 2011-12-30 | 2012-07-11 | 河海大学常州校区 | Plant growth condition monitoring system based on compute vision and internet of things |
WO2014134993A1 (en) * | 2013-03-04 | 2014-09-12 | 北京联想软件有限公司 | Target object information acquisition method and electronic device |
CN105526887A (en) * | 2016-02-23 | 2016-04-27 | 东华大学 | Non-contact type fabric surface roughness test method and test device |
Cited By (16)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109190642A (en) * | 2018-09-04 | 2019-01-11 | 华中科技大学 | The method for extracting surface characteristics using high-order Gauss regression filtering and Radon transformation |
CN109408888B (en) * | 2018-11-27 | 2024-01-05 | 广东工业大学 | Roughness calculation method of two-dimensional cutting surface, computer-readable storage medium and terminal |
CN109408888A (en) * | 2018-11-27 | 2019-03-01 | 广东工业大学 | A kind of roughness calculation method, computer readable storage medium and the terminal on two-dimensional cutting surface |
CN109357636A (en) * | 2018-12-10 | 2019-02-19 | 电子科技大学 | A kind of phase amplification calculation method based on black part structure light scan |
CN109357636B (en) * | 2018-12-10 | 2019-12-17 | 电子科技大学 | phase amplification calculation method based on black piece structured light scanning |
CN110211100A (en) * | 2019-05-20 | 2019-09-06 | 浙江大学 | A kind of foot measurement method of parameters based on image |
CN112050756A (en) * | 2020-09-04 | 2020-12-08 | 南通大学 | Rock ore slice and resin target surface flatness recognition processing method |
CN112050756B (en) * | 2020-09-04 | 2022-05-06 | 南通大学 | Rock ore slice and resin target surface flatness recognition processing method |
CN113701632A (en) * | 2021-09-01 | 2021-11-26 | 威海北洋电气集团股份有限公司 | Thread detection method based on difference value |
CN113701632B (en) * | 2021-09-01 | 2024-02-13 | 威海北洋电气集团股份有限公司 | Thread detection method based on difference value |
CN114046749A (en) * | 2021-10-26 | 2022-02-15 | 刘红霞 | Method and system for detecting roughness of point-shaped pit joint surface of precast concrete member |
CN114331923A (en) * | 2022-03-11 | 2022-04-12 | 中国空气动力研究与发展中心低速空气动力研究所 | Improved Canny algorithm-based bubble contour extraction method in ice structure |
CN114677340A (en) * | 2022-03-14 | 2022-06-28 | 上海第二工业大学 | Concrete surface roughness detection method based on image edge |
CN114677340B (en) * | 2022-03-14 | 2024-05-24 | 上海第二工业大学 | Concrete surface roughness detection method based on image edge |
CN117523502A (en) * | 2024-01-08 | 2024-02-06 | 东莞市杰瑞智能科技有限公司 | Urban road rubbish intelligent monitoring system based on machine vision |
CN117523502B (en) * | 2024-01-08 | 2024-03-19 | 东莞市杰瑞智能科技有限公司 | Urban road rubbish intelligent monitoring system based on machine vision |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106767564A (en) | A kind of detection method for being applied to phone housing surface roughness | |
CN109141232B (en) | Online detection method for disc castings based on machine vision | |
CN110286124B (en) | Machine vision-based refractory brick measuring system | |
CN107341802B (en) | Corner sub-pixel positioning method based on curvature and gray scale compounding | |
CN104112269B (en) | A kind of solar battery laser groove parameter detection method and system based on machine vision | |
CN102441581A (en) | Machine vision-based device and method for online detection of structural steel section size | |
Chen et al. | Automated inspection of engineering ceramic grinding surface damage based on image recognition | |
CN111583114B (en) | Automatic measuring device and measuring method for pipeline threads | |
CN115222733A (en) | Metal component surface defect detection method based on image recognition | |
CN109284718B (en) | Inspection robot-oriented variable-view-angle multi-instrument simultaneous identification method | |
CN106824806A (en) | The detecting system and detection method of the low module plastic gear based on machine vision | |
Deng et al. | Defect detection of bearing surfaces based on machine vision technique | |
CN110717909A (en) | Metal surface scratch detection method and device | |
CN115063407A (en) | Scratch and crack identification method for annular copper gasket | |
Karnawat et al. | Turbidity detection using image processing | |
CN110728286B (en) | Abrasive belt grinding material removal rate identification method based on spark image | |
CN115063620A (en) | Bit layering-based Roots blower bearing wear detection method | |
Elhamshary et al. | Surface Roughness Measurement Using Light Sectioning Method and Computer Vision Techniques. | |
CN114549403B (en) | Method for detecting intelligent precise geometrical circle centers of multiple monomers of side section of mechanical part | |
CN115761257A (en) | Method and device for detecting angle of part based on computer vision technology | |
Li et al. | Vehicle seat detection based on improved RANSAC-SURF algorithm | |
CN113591923A (en) | Engine rocker arm part classification method based on image feature extraction and template matching | |
Bonnot et al. | Machine vision system for surface inspection on brushed industrial parts | |
Wang et al. | Surface Defect Recognition and Invalidation Judgment of Remanufactured Gears Based on Machine Vision | |
Karegoudra et al. | Estimating the Surface Roughness of a Drilled Surface Using the Edge Detection Technique |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170531 |
|
RJ01 | Rejection of invention patent application after publication |