CN106530291A - Polished surface detection method and system based on image processing - Google Patents
Polished surface detection method and system based on image processing Download PDFInfo
- Publication number
- CN106530291A CN106530291A CN201610968488.6A CN201610968488A CN106530291A CN 106530291 A CN106530291 A CN 106530291A CN 201610968488 A CN201610968488 A CN 201610968488A CN 106530291 A CN106530291 A CN 106530291A
- Authority
- CN
- China
- Prior art keywords
- gray
- pixel
- image
- scale
- row
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
- G06T7/0006—Industrial image inspection using a design-rule based approach
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10056—Microscopic image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30164—Workpiece; Machine component
Landscapes
- Engineering & Computer Science (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a polished surface detection method based on image processing, relating to the field of image processing. The method comprises a step of collecting an image comprising a target object surface, carrying out gray processing on the image, dividing a plurality of areas, and taking the gray average value of the areas as the area pixel gray value, a step of substituting the gray of an original image pixel point by Gray Phi(i) according to a threshold interval Phi(i), and a step of collecting image gray change information, obtaining pixel gray fluctuation value, and marking work piece polishing as unqualified if the fluctuation value is larger than a set threshold. In addition, the invention also discloses a polished surface detection system based on image processing. A manual operation is not needed, the system is easy to be integrated with an industrial production line, the labor cost is saved, the surface detection judgment speed is fast, and the realization of industrialization is easy.
Description
Technical field
The present invention relates to Surface testing field, more particularly to a kind of polished surface detection method based on image procossing and
Its system.
Background technology
Material technology is one of key technology for giving priority to 21 century, modern machines manufacturing industry and other many industries
Development is put forward higher requirement to metal material.The development of metal material science and technology is rapider.Wherein automobile industry table
Face smoothness requirements are very strict, at present for the Surface testing of polishing process, mostly artificial to operate, and operation is relatively independent,
Human cost is very high.
Surface of the work polishing is conducive to improving Workpiece Machining Accuracy, improves surface of the work aesthetic feeling etc., is the weight of work piece production
Want step.Even if however, improving the precision of polishing process, can still there is pit and projection in surface of the work.So, in polishing precision
In detection, the rough degree of surface of the work is only needed to can determine that workpiece is qualified within technique management and control.
Conventional images process Surface testing, and mainly surface of the work feature is identified, and its detection cost time is longer;
And in industrial producing line, each procedure time is compact, it is impossible to take a long time on every workpiece;In a word, existing skill
Whether art can not evaluate polished surface in time in polished surface domain variability qualified.
The content of the invention
In view of the drawbacks described above of prior art, the technical problem to be solved is to provide a kind of based at image
The polished surface detection method of reason and its system, solve prior art using artificial operation, and operation is relatively independent, and human cost is high
The problems such as.Meanwhile, the present invention also solves existing employing image procossing and obtains the slow-footed problem of Surface testing, there is provided Yi Zhongji
In the method and system of the quick detection workpiece polished surface of image procossing.
For achieving the above object, the invention provides a kind of polished surface detection method based on image procossing, including such as
Lower step:
Step S1, original image of the collection comprising target object surface;
Step S2, by the original image gray processing, and the original image is divided into into some regions;Take described in each
The average gray in region is used as the area pixel point gray value;
Step S3, I gray threshold interval Φ of settingi, by each pixel gray level difference of the image Jing after the process of step S2
Replace with each gray threshold interval ΦiCorresponding gray scale setting valueWherein, the I is positive integer, the i satisfactions 1
≤i≤I;
Step S4, the collection horizontal pixel gray scale of image Jing after the process of step S3 rises number Hup, under horizontal pixel gray scale
Drop number Hdown;The longitudinal pixel grey scale for gathering the image Jing after the process of step S3 rises number Zup, longitudinal pixel grey scale declines
Number ZdownAnd Jing step S3 process after image pixel total number Sum;The Sum is positive integer;The Hup、Hdown、Zup、Zdown
For natural number;
Step S5, the average fluctuation amplitude F of acquisition gray scale, it is describedThe Grayj
For each pixel gray value of the image Jing after the process of step S3;It is describedFor the average gray of the image Jing after the process of step S3
Value;The j meets 1≤j≤Sum;
Step S6, acquisition pixel grey scale undulating value E, if E>ETH, then send sound and light alarm;It is described
The ETHSpan is 0.01≤ETH≤10;The ETHFor pixel grey scale fluctuation threshold.
In the technical scheme, surface of the work is detected by image procossing, save human cost, improve workpiece inspection
Efficiency is surveyed, workpiece sensing speed is improved.Meanwhile, in step s 2, carry out gray processing to original image, zoning, and with each
On the one hand the average gray in region reduces image processing data and further improves detection speed, separately as the area grayscale value
On the one hand, multiple pixels are constituted into a region, slight flaws can be hidden by the average region, not to slight flaws or
Defect within technique management and control is passed judgment on, and can improve the accuracy and speed of detection.In step s3, set gray scale threshold
Value is interval, 255 GTGs is reduced to 8,32 or other numbers, image processing speed can be improved and be eliminated workpiece
The negligible nibs in surface.It is noted that what step S2 considered emphatically is workpiece on Spatial Dimension or size
Slight flaws, when the width dimensions of pit are simultaneously little, then ignore pit;And step S3 focuses on the workpiece for considering in flaw degree,
When the depth dimensions of pit is unsatisfactory for workpiece grayscale difference, then ignore the pit.
Additionally, in the technical scheme, extracting grey scale change information, numerical operation is directly carried out, obtain pixel grey scale ripple
It is dynamic, it is quick to obtain the whether qualified information of workpiece, improve surface of the work detection speed and precision.Pass through
Average fluctuation amplitude F of the pixel relative to average gray is obtained, then is passed through
Obtain pixel grey scale undulating value, the physical significance of the formula is, the pixel grey scale undulating value E and average fluctuation amplitude F of gray scale into
Direct ratio, the geometrical mean of the number that fluctuates with transverse and longitudinalBe directly proportional, with pixel total number Sum into
Inverse ratio.In a word, it is beneficial in that using step S4, S5, S6, fast and accurately obtains surface of the work testing result.
Furthermore, the collection horizontal pixel gray scale rises number HupThe step of include:
Step S411:Gather in m rows, latter pixel gray value Gray(m,n)More than previous pixel gray value
Gray(m,n-1)Single file horizontal pixel rise number Hup_m;
Step S412:Number H is risen to the single file horizontal pixel of each rowup_mSummation, obtains in horizontal pixel gray scale
Rise number Hup;It is described
The collection horizontal pixel gray scale declines number HdownThe step of include:
Step S421:Gather in m rows, latter pixel gray value Gray(m,n)Less than previous pixel gray value
Gray(m,n-1)Single file horizontal pixel decline number Hdown_m;
Step S422:Number H is declined to the single file horizontal pixel of each rowdown_mSummation, obtains under horizontal pixel gray scale
Drop number Hdown;It is described
Wherein, the 1≤m≤M, 2≤n≤N;The M is the total line number of image, and the N is the total columns of image.
In the technical scheme, by the relation that rises or falls for obtaining pixel and neighboring pixel, statistics gray scale rise and
Lower drop data, contributes to the numerical operation of subsequent step, improves accuracy of detection.
Furthermore, the longitudinal pixel grey scale of the collection rises number ZupThe step of include:
Step S431:In the n-th row of collection, latter pixel gray value Gray(m,n)More than previous pixel gray value
Gray(m-1,n)Single-row longitudinal pixel rise number Zup_n;
Step S432:Number Z is risen to described single-row longitudinal pixel of each rowup_nSummation, obtains on longitudinal pixel grey scale
Rise number Zup;It is described
Collection longitudinal direction pixel grey scale declines number ZdownThe step of include:
Step S441:In the n-th row of collection, latter pixel gray value Gray(m,n)Less than previous pixel gray value
Gray(m-1,n)Single-row longitudinal pixel decline number Zdown_n;
Step S442:Number Z is declined to the single file longitudinal direction pixel of each rowdown_nSummation, obtains under longitudinal pixel grey scale
Drop number Zdown;It is described
Wherein, the 2≤m≤M, 1≤n≤N;The M is the total line number of image, and the N is the total columns of image.
In the technical scheme, by the relation that rises or falls for obtaining pixel and neighboring pixel, statistics gray scale rise and
Lower drop data, contributes to the numerical operation of subsequent step, improves accuracy of detection.
Furthermore, in step S1, the original image comprising target object surface is MIcrosope image;
In step S6, pixel grey scale fluctuation threshold E is set according to microscopical multiplication factor βTH;The ETH=0.01 β;Institute
State 0.01≤β≤12000.
In the technical scheme, impact of the microscope magnification to surface of the work image is fully taken into account, setting is different
Fluctuation threshold ETH, improve the accuracy of Surface testing.It is noted that microscope magnification is bigger, image detail shows
Show that about substantially the average fluctuation amplitude of gray scale and pixel grey scale undulating value for showing as image is also bigger, so need setting higher
Fluctuation threshold, carry out the whether qualified detection of calibration of workpieces.
In view of the defect of prior art, the present invention also provides a kind of polished surface detecting system based on image procossing,
Including:
Image capture module, for original image of the collection comprising target object surface;
The original image for by the original image gray processing, and is divided into some regions by picture portion module;
The average gray in each region is taken as the area pixel point gray value;
Greyscale transformation module, for setting I gray threshold interval Φi, will scheme Jing after the process of described image division module
Each pixel gray level of picture replaces with each gray threshold interval Φ respectivelyiCorresponding gray scale setting valueWherein, it is described
I is positive integer, and the i meets 1≤i≤I;
Variation of image grayscale acquisition module, for gathering the horizontal pixel ash of image after the greyscale transformation resume module described in
Degree rises number Hup, horizontal pixel gray scale decline number Hdown;The longitudinal direction of image after collection greyscale transformation resume module described in
Pixel grey scale rises number Zup, longitudinal pixel grey scale decline number ZdownAnd image slices after the greyscale transformation resume module described in
Plain total number Sum;The Sum is positive integer;The Hup、Hdown、Zup、ZdownFor natural number;
Gradation of image fluctuation solves module, for obtaining the average fluctuation amplitude F of gray scale, described
The GrayjFor each pixel gray value of image after the greyscale transformation resume module described in;It is describedThe ash described in Jing
The average gray value of image after degree conversion module process;The j meets 1≤j≤Sum;
Workpiece certified products determination module, for obtaining pixel grey scale undulating value E, if E>ETH, then send sound and light alarm;It is describedThe ETHSpan is 0.01≤ETH≤10;The ETHFor pixel
Gray scale fluctuation threshold.
In the technical scheme, surface of the work is detected by image procossing, save human cost, improve workpiece inspection
Efficiency is surveyed, workpiece sensing speed is improved.Meanwhile, in step s 2, carry out gray processing to original image, zoning, and with each
On the one hand the average gray in region reduces image processing data and further improves detection speed, separately as the area grayscale value
On the one hand, multiple pixels are constituted into a region, slight flaws can be hidden by the average region, not to slight flaws or
Defect within technique management and control is passed judgment on, and can improve the accuracy and speed of detection.In step s3, set gray scale threshold
Value is interval, 255 GTGs is reduced to 8,32 or other numbers, image processing speed can be improved and be eliminated workpiece
The negligible nibs in surface.Additionally, extracting grey scale change information, numerical operation is directly carried out, obtains pixel grey scale fluctuation,
It is quick to obtain the whether qualified information of workpiece, improve surface of the work detection speed.
Furthermore, described image grey scale change acquisition module, including:Laterally gray scale rises number solution unit, horizontal stroke
Decline number solution unit, longitudinal gray scale rising number solution unit and longitudinal gray scale decline number to gray scale and solve unit.
In the technical scheme, by the relation that rises or falls for obtaining pixel and neighboring pixel, statistics gray scale rise and
Lower drop data, contributes to the numerical operation of subsequent step, improves accuracy of detection.
Furthermore, the horizontal gray scale rises number solution unit, is configured to:
Gather in m rows, latter pixel gray value Gray(m,n)More than previous pixel gray value Gray(m,n-1)
Single file horizontal pixel rise number Hup_m;
Number H is risen to the single file horizontal pixel of each rowup_mSummation, obtains horizontal pixel gray scale and rises number Hup;
It is described
The horizontal gray scale declines number and solves unit, is configured to:
Gather in m rows, latter pixel gray value Gray(m,n)Less than previous pixel gray value Gray(m,n-1)
Single file horizontal pixel decline number Hdown_m;
Number H is declined to the single file horizontal pixel of each rowdown_mSummation, obtains horizontal pixel gray scale and declines number
Hdown;It is described
Wherein, the 1≤m≤M, 2≤n≤N;The M is the total line number of image, and the N is the total columns of image.
In the technical scheme, by the relation that rises or falls for obtaining pixel and neighboring pixel, statistics gray scale rise and
Lower drop data, contributes to the numerical operation of subsequent step, improves accuracy of detection.
Furthermore, longitudinal gray scale rises number solution unit, is configured to:
In the n-th row of collection, latter pixel gray value Gray(m,n)More than previous pixel gray value Gray(m-1,n)
Single-row longitudinal pixel rise number Zup_n;
Number Z is risen to described single-row longitudinal pixel of each rowup_nSummation, obtains longitudinal pixel grey scale and rises number Zup;
It is described
Longitudinal gray scale declines number and solves unit, is configured to:
In the n-th row of collection, latter pixel gray value Gray(m,n)Less than previous pixel gray value Gray(m-1,n)
Single-row longitudinal pixel decline number Zdown_n;
Number Z is declined to the single file longitudinal direction pixel of each rowdown_nSummation, obtains longitudinal pixel grey scale and declines number
Zdown;It is described
Wherein, the 2≤m≤M, 1≤n≤N;The M is the total line number of image, and the N is the total columns of image.
In the technical scheme, by the relation that rises or falls for obtaining pixel and neighboring pixel, statistics gray scale rise and
Lower drop data, contributes to the numerical operation of subsequent step, improves accuracy of detection.
Furthermore, the image of described image acquisition module collection is MIcrosope image;The workpiece certified products judge
Module, also including gray scale fluctuation threshold setup unit;The gray scale fluctuation threshold setup unit, is configured to:According to microscope
Multiplication factor β setting pixel grey scale fluctuation threshold ETH;The ETH=0.01 β;0.01≤β≤12000.
In the technical scheme, impact of the microscope magnification to surface of the work image is fully taken into account, setting is different
Fluctuation threshold ETH, improve the accuracy of Surface testing.It is noted that microscope magnification is bigger, image detail shows
Show that about substantially the average fluctuation amplitude of gray scale and pixel grey scale undulating value for showing as image is also bigger, so need setting higher
Fluctuation threshold, carry out the whether qualified detection of calibration of workpieces.
The invention has the beneficial effects as follows:It is of the invention surface of the work to be detected by image procossing, human cost is saved,
Workpiece sensing efficiency is improved, workpiece sensing speed is improved.Additionally, the present invention carries out gray processing to original image, zoning, and with
The average gray of regional is as the area grayscale value, then interval in setting gray threshold and carry out depression of order to gray scale,
Improve detection speed and precision.Finally, the present invention fully takes into account impact of the microscope magnification to surface of the work image, if
Fixed different fluctuation threshold ETH, improve the accuracy of Surface testing.
Description of the drawings
Fig. 1 is the schematic flow sheet of the embodiment of the invention;
Fig. 2 is the structural representation of image-region division in the embodiment of the invention;
Fig. 3 is the explanation schematic diagram of image-region division and greyscale transformation in the embodiment of the invention;
Fig. 4 is the non-switched image single file grey scale change curve of gray value in the embodiment of the invention;
Fig. 5 is the image single file grey scale change curve in the embodiment of the invention after grayvalue transition;
Fig. 6 is the structural representation of another specific embodiment of the invention.
Specific embodiment
The invention will be further described with reference to the accompanying drawings and examples:
As shown in figure 1, in the first embodiment, there is provided a kind of polished surface detection method based on image procossing, including
Following steps:
Step S1, original image of the collection comprising target object surface;
Step S2, by the original image gray processing, and the original image is divided into into some regions;Take described in each
The average gray in region is used as the area pixel point gray value;
Step S3, I gray threshold interval Φ of settingi, by each pixel gray level difference of the image Jing after the process of step S2
Replace with each gray threshold interval ΦiCorresponding gray scale setting valueWherein, the I is positive integer, the i satisfactions 1
≤i≤I;
Step S4, the collection horizontal pixel gray scale of image Jing after the process of step S3 rises number Hup, under horizontal pixel gray scale
Drop number Hdown;The longitudinal pixel grey scale for gathering the image Jing after the process of step S3 rises number Zup, longitudinal pixel grey scale declines
Number ZdownAnd Jing step S3 process after image pixel total number Sum;The Sum is positive integer;The Hup、Hdown、Zup、Zdown
For natural number;
Step S5, the average fluctuation amplitude F of acquisition gray scale, it is describedThe Grayj
For each pixel gray value of the image Jing after the process of step S3;It is describedFor the average gray of the image Jing after the process of step S3
Value;The j meets 1≤j≤Sum;
Step S6, acquisition pixel grey scale undulating value E, if E>ETH, then send sound and light alarm;It is described
The ETHSpan is 0.01≤ETH≤10;The ETHFor pixel grey scale fluctuation threshold.
Below by embodiment, it is described in detail for each step of the invention.
As shown in figures 1 to 6, in the first embodiment, there is provided a kind of polished surface detection method based on image procossing, wrap
Include following steps:
Step S1, original image of the collection comprising target object surface;
It is noted that in disclosure example, image directly can be gathered by CCD/CMOS cameras, it is also possible to logical
Microscope amplification is crossed, and gathers the image after microscope amplifies.
Preferably, the image comprising target object surface is MIcrosope image;
It should be noted is that:The surface inspecting method that the embodiment of the present disclosure is provided, can shoot under microscope
The image shot under image, or conventional common multiple, the embodiment of the present disclosure are not construed as limiting to this.Need a bit for illustrating
It is in the present embodiment, to be only illustrated so that shooting image under the microscope carries out surface of the work detection as an example.At which
In its possible embodiment, image procossing is carried out to the conventional image for shooting lower acquisition, the testing result of surface of the work is obtained.
Shooting image under microscope, it is possible to obtain more surface of the work details, improves Surface testing precision, which increases certainly
System cost.In actual applications, can be selected according to the actual requirements, the disclosure is not construed as limiting to this.
Step S2, by the original image gray processing, and the original image is divided into into some regions;Take described in each
The average gray in region is used as the area pixel point gray value;
In implementing one, some regions are divided into by image.It is such as Fig. 2, exemplary, with 3 × 3 pixel of original image for stroke
The single region divided, and using the average gray value of 3 × 3 pixels as the gray value of new images.
It is noted that visually observing polishing surface of the work from macroscopic view, surface of the work is smooth.And pass through microscopic observation,
Surface of the work presents uneven.Preferably, MIcrosope image is obtained using microscope, surface of the work is presented obvious hollow
Phenomenon.Meanwhile, workpiece is mostly opaque, and light source is projected from microscopical object lens, and Jing surface of the works are reflective, through microscope
System, is received by the ccd image collection of eyepiece.Now, the gray value on MIcrosope image, can preferably weigh workpiece table
Face planarization, and the pixel gray value height of eminence is shown as, the pixel gray value of lower is low.
Additionally, the pixel in regional is averaged, on the one hand can reduce image intensity value rise number and
Decline number, reduce and calculate data volume, improve processing speed;On the other hand, the flaw in smaller area and its gray scale are fluctuated
Filter out, ignore nibs.
Step S3, I gray threshold interval Φ of settingi, by each pixel gray level difference of the image Jing after the process of step S2
Replace with each gray threshold interval ΦiCorresponding gray scale setting valueWherein, the I is positive integer, the i satisfactions 1
≤i≤I;
In the present invention, the gray value of image is believed relative to the height of average height in fact, include surface of the work
Breath, gray threshold interval ΦiSetting can set based on experience value;Can also be obtained by experiment test.Wherein, threshold zone
Between can be with non-linear setting, it would however also be possible to employ linear wide setting.
Exemplary, such as table 1, the gray value of 255 ranks is averagely divided to into some threshold interval Φ in the present embodimenti
In, and with each threshold interval ΦiIntermediate value as the interval gray value Grayi.Table 2 is the image I in some regions in certain row
(x, y) and according to threshold interval ΦiReplace the tables of data of image J (x, y) after each pixel gray level;It is noted that
Due to Jing step S2 process, in regional, pixel gray value is the same, so only gray value is provided by region in table 2.In figure
4th, the curve of data before and after changing is given in Fig. 5.
Threshold interval Φ in table 1, an embodimentiWith gray scale GrayiTransformational relation table
Interval numbering i | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 |
Threshold interval Φi | [0,31] | [32,63] | [64,95] | [96,127] | [128,159] | [160,191] | [192,223] | [224,255] |
Grayi | 16 | 48 | 80 | 112 | 144 | 176 | 208 | 240 |
Table 2, based on threshold interval ΦiGradation conversion example
Numbering | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 |
I (x, y) image intensity value | 127 | 132 | 131 | 126 | 120 | 80 | 83 | 87 | 82 | 96 | 83 | 81 |
J (x, y) image intensity value | 112 | 112 | 112 | 112 | 112 | 80 | 80 | 80 | 80 | 80 | 80 | 80 |
Numbering | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 23 | 24 |
I (x, y) image intensity value | 93 | 110 | 118 | 145 | 155 | 159 | 167 | 172 | 185 | 188 | 190 | 195 |
J (x, y) image intensity value | 80 | 80 | 112 | 144 | 144 | 144 | 144 | 144 | 176 | 176 | 176 | 176 |
Numbering | 25 | 26 | 27 | 28 | 29 | 30 | 31 | 32 | 33 | 34 | 35 | |
I (x, y) image intensity value | 183 | 186 | 181 | 196 | 186 | 172 | 150 | 145 | 122 | 100 | 111 | |
J (x, y) image intensity value | 176 | 176 | 176 | 176 | 176 | 144 | 144 | 144 | 112 | 80 | 112 |
It is noted that what step S2 considered emphatically is slight flaws of the workpiece on Spatial Dimension or size, when recessed
The width dimensions in hole are simultaneously little, then ignore pit;And step S3 focuses on the workpiece for considering in flaw degree, when the depth gauge rule of pit
It is very little to be unsatisfactory for workpiece grayscale difference, then ignore the pit.As shown in figure 3, the width dimensions of pit 203 are less, ignore pit 203
Get off;204 width dimensions of pit are larger, and depth is larger, and the flaw is not ignored.Although 301 width dimensions of pit are larger,
But depth is shallower, and it is not ignored;302 width dimensions of pit are larger, and depth dimensions also meets, then should not ignore.
Step S4, step S4, the collection horizontal pixel gray scale of image Jing after the process of step S3 rises number Hup, horizontal pixel
Gray scale declines number Hdown;The longitudinal pixel grey scale for gathering the image Jing after the process of step S3 rises number Zup, longitudinal pixel grey scale
Decline number ZdownAnd Jing step S3 process after image pixel total number Sum;The Sum is positive integer;The Hup、Hdown、
Zup、ZdownFor natural number;Wherein,
Collection horizontal pixel gray scale rises number HupThe step of include:
Step S411:Gather in m rows, latter pixel gray value Gray(m,n)More than previous pixel gray value
Gray(m,n-1)Single file horizontal pixel rise number Hup_m;
Step S412:Number H is risen to the single file horizontal pixel of each rowup_mSummation, obtains in horizontal pixel gray scale
Rise number Hup;It is described
Collection horizontal pixel gray scale declines number HdownThe step of include:
Step S421:Gather in m rows, latter pixel gray value Gray(m,n)Less than previous pixel gray value
Gray(m,n-1)Single file horizontal pixel decline number Hdown_m;
Step S422:Number H is declined to the single file horizontal pixel of each rowdown_mSummation, obtains under horizontal pixel gray scale
Drop number Hdown;It is described
Wherein, the 2≤m≤M, 1≤n≤N;The M is the total line number of described image J (x, y), and the N is described image J
(x, y) total columns.
Exemplary, as shown in figure 5, counting to the data variation information of image J (x, y) m rows, obtain single file horizontal
Rise number H to pixelup_m=4, single file horizontal pixel declines number Hdown_m=4.
Collection longitudinal direction pixel grey scale rises number ZupThe step of include:
Step S431:In the n-th row of collection, latter pixel gray value Gray(m,n)More than previous pixel gray value
Gray(m-1,n)Single-row longitudinal pixel rise number Zup_n;
Step S432:Number Z is risen to described single-row longitudinal pixel of each rowup_nSummation, obtains on longitudinal pixel grey scale
Rise number Zup;It is described
Collection longitudinal direction pixel grey scale declines number ZdownThe step of include:
Step S441:In the n-th row of collection, latter pixel gray value Gray(m,n)Less than previous pixel gray value
Gray(m-1,n)Single-row longitudinal pixel decline number Zdown_n;
Step S442:Number Z is declined to the single file longitudinal direction pixel of each rowdown_nSummation, obtains under longitudinal pixel grey scale
Drop number Zdown;It is described
Wherein, the 1≤m≤M, 2≤n≤N;The M is the total line number of described image J (x, y), and the N is described image J
(x, y) total columns.
Step S5, the average fluctuation amplitude F of acquisition gray scale, it is describedThe Grayj
For each pixel gray value of the image Jing after the process of step S3;It is describedFor the average gray of the image Jing after the process of step S3
Value;The j meets 1≤j≤Sum;Wherein,
Step S6, acquisition pixel grey scale undulating value E, if E>ETH, then send sound and light alarm;It is described
Pixel grey scale undulating value E is obtained, if E>ETH, then mark the workpiece to be finished to unqualified, and send sound and light alarm;
If E≤ETH, then mark the workpiece polishing qualified.It is describedThe ETH
Span is 0.01≤ETH≤10;The ETHFor pixel grey scale fluctuation threshold.
Preferably, the step of the present embodiment in S1, image H (x, y) comprising target object surface is MIcrosope image;
In step s 6, pixel grey scale fluctuation threshold E is set according to microscopical multiplication factor βTH;The ETH=0.01 β.
It is noted that in another alternative embodiment of the invention, being taken the photograph by common CCD or CMOS during image H (x, y)
Obtain as head shoots, ETH=0.01.
As shown in fig. 6, in second embodiment of the invention, there is provided a kind of polished surface detection system based on image procossing
System, including:
Image capture module 1, for original image of the collection comprising target object surface;
The original image for by the original image gray processing, and is divided into some areas by picture portion module 2
Domain;The average gray in each region is taken as the area pixel point gray value;
Greyscale transformation module 3, for setting I gray threshold interval Φi, by Jing after described image division module 2 is processed
Each pixel gray level of image replaces with each gray threshold interval Φ respectivelyiCorresponding gray scale setting valueWherein, institute
I is stated for positive integer, the i satisfactions 1≤i≤I;
Variation of image grayscale acquisition module 4, for gathering the horizontal pixel of image after the process of greyscale transformation module 3 described in
Gray scale rises number Hup, horizontal pixel gray scale decline number Hdown;After collection process of greyscale transformation module 3 described in, image is vertical
Rise number Z to pixel grey scaleup, longitudinal pixel grey scale decline number ZdownAnd scheme after described in, greyscale transformation module 3 is processed
As pixel total number Sum;The Sum is positive integer;The Hup、Hdown、Zup、ZdownFor natural number;
Gradation of image fluctuation solves module 5, for obtaining the average fluctuation amplitude F of gray scale, described
The GrayjEach pixel gray value of image after processing for the greyscale transformation module 3 described in;It is describedThe ash described in Jing
The average gray value of image after the degree process of conversion module 3;The j meets 1≤j≤Sum;
Workpiece certified products determination module 6, for obtaining pixel grey scale undulating value E, if E>ETH, then send sound and light alarm;Institute
StateThe ETHSpan is 0.01≤ETH≤10;The ETHIt is picture
Plain gray scale fluctuation threshold.
In the present embodiment, variation of image grayscale acquisition module 4, including:Laterally gray scale rises number and solves unit, laterally
Gray scale declines number solution unit, longitudinal gray scale rising number solution unit and longitudinal gray scale decline number and solves unit.
In the present embodiment, it is characterised in that the horizontal gray scale rises number and solves unit, is configured to:
Gather in m rows, latter pixel gray value Gray(m,n)More than previous pixel gray value Gray(m,n-1)
Single file horizontal pixel rise number Hup_m;
Number H is risen to the single file horizontal pixel of each rowup_mSummation, obtains horizontal pixel gray scale and rises number Hup;
It is described
The horizontal gray scale declines number and solves unit, is configured to:
Gather in m rows, latter pixel gray value Gray(m,n)Less than previous pixel gray value Gray(m,n-1)
Single file horizontal pixel decline number Hdown_m;
Number H is declined to the single file horizontal pixel of each rowdown_mSummation, obtains horizontal pixel gray scale and declines number
Hdown;It is described
Wherein, the 1≤m≤M, 2≤n≤N;The M is the total line number of image, and the N is the total columns of image.
In the present embodiment, longitudinal gray scale rises number solution unit, is configured to:
In the n-th row of collection, latter pixel gray value Gray(m,n)More than previous pixel gray value Gray(m-1,n)
Single-row longitudinal pixel rise number Zup_n;
Number Z is risen to described single-row longitudinal pixel of each rowup_nSummation, obtains longitudinal pixel grey scale and rises number Zup;
It is described
Longitudinal gray scale declines number and solves unit, is configured to:
In the n-th row of collection, latter pixel gray value Gray(m,n)Less than previous pixel gray value Gray(m-1,n)
Single-row longitudinal pixel decline number Zdown_n;
Number Z is declined to the single file longitudinal direction pixel of each rowdown_nSummation, obtains longitudinal pixel grey scale and declines number
Zdown;It is described
Wherein, the 2≤m≤M, 1≤n≤N;The M is the total line number of image, and the N is the total columns of image.
In the present embodiment, the image of the collection of described image acquisition module 1 is MIcrosope image;The workpiece certified products are sentenced
Cover half block 6, also including gray scale fluctuation threshold setup unit;The gray scale fluctuation threshold setup unit, is configured to:According to micro-
Multiplication factor β setting pixel grey scale fluctuation threshold E of mirrorTH;The ETH=0.01 β;0.01≤β≤12000.
The preferred embodiment of the present invention described in detail above.It should be appreciated that one of ordinary skill in the art without
Need creative work just can make many modifications and variations with design of the invention.Therefore, all technologies in the art
Personnel are available by logical analysis, reasoning, or a limited experiment on the basis of existing technology under this invention's idea
Technical scheme, all should be in the protection domain being defined in the patent claims.
Claims (9)
1. a kind of polished surface detection method based on image procossing, it is characterised in that comprise the steps:
Step S1, original image of the collection comprising target object surface;
Step S2, by the original image gray processing, and the original image is divided into into some regions;Take each region
Average gray as the area pixel point gray value;
Step S3, I gray threshold interval Φ of settingi, each pixel gray level of the image Jing after the process of step S2 is replaced with respectively
Each gray threshold interval ΦiCorresponding gray scale setting valueWherein, the I is positive integer, 1≤i of the i satisfactions≤
I;
Step S4, the collection horizontal pixel gray scale of image Jing after the process of step S3 rises number Hup, horizontal pixel gray scale decline
Number Hdown;The longitudinal pixel grey scale for gathering the image Jing after the process of step S3 rises number Zup, longitudinal pixel grey scale decline number
ZdownAnd Jing step S3 process after image pixel total number Sum;The Sum is positive integer;The Hup、Hdown、Zup、ZdownFor
Natural number;
Step S5, the average fluctuation amplitude F of acquisition gray scale, it is describedThe GrayjFor Jing
Each pixel gray value of image after the process of step S3;It is describedFor the average gray value of the image Jing after the process of step S3;
The j meets 1≤j≤Sum;
Step S6, acquisition pixel grey scale undulating value E, if E>ETH, then send sound and light alarm;It is described
The ETHSpan is 0.01≤ETH≤10;The ETHFor pixel grey scale fluctuation threshold.
2. a kind of polished surface detection method based on image procossing as claimed in claim 1, it is characterised in that
The collection horizontal pixel gray scale rises number HupThe step of include:
Step S411:Gather in m rows, latter pixel gray value Gray(m,n)More than previous pixel gray value
Gray(m,n-1)Single file horizontal pixel rise number Hup_m;
Step S412:Number H is risen to the single file horizontal pixel of each rowup_mSummation, obtains horizontal pixel gray scale and rises number
Hup;It is described
The collection horizontal pixel gray scale declines number HdownThe step of include:
Step S421:Gather in m rows, latter pixel gray value Gray(m,n)Less than previous pixel gray value
Gray(m,n-1)Single file horizontal pixel decline number Hdown_m;
Step S422:Number H is declined to the single file horizontal pixel of each rowdown_mSummation, obtains the decline of horizontal pixel gray scale
Number Hdown;It is described
Wherein, the 1≤m≤M, 2≤n≤N;The M is the total line number of image, and the N is the total columns of image.
3. a kind of polished surface detection method based on image procossing as claimed in claim 1, it is characterised in that
Collection longitudinal direction pixel grey scale rises number ZupThe step of include:
Step S431:In the n-th row of collection, latter pixel gray value Gray(m,n)More than previous pixel gray value
Gray(m-1,n)Single-row longitudinal pixel rise number Zup_n;
Step S432:Number Z is risen to described single-row longitudinal pixel of each rowup_nSummation, obtains longitudinal pixel grey scale and rises number
Zup;It is described
Collection longitudinal direction pixel grey scale declines number ZdownThe step of include:
Step S441:In the n-th row of collection, latter pixel gray value Gray(m,n)Less than previous pixel gray value
Gray(m-1,n)Single-row longitudinal pixel decline number Zdown_n;
Step S442:Number Z is declined to the single file longitudinal direction pixel of each rowdown_nSummation, obtains longitudinal pixel grey scale decline
Number Zdown;It is described
Wherein, the 2≤m≤M, 1≤n≤N;The M is the total line number of image, and the N is the total columns of image.
4. a kind of polished surface detection method based on image procossing as claimed in claim 1, it is characterised in that:
In step S1, the original image comprising target object surface is MIcrosope image;
In step S6, pixel grey scale fluctuation threshold E is set according to microscopical multiplication factor βTH;The ETH=0.01 β;
0.01≤β≤12000.
5. a kind of polished surface detecting system based on image procossing, it is characterised in that include:
Image capture module, for original image of the collection comprising target object surface;
The original image for by the original image gray processing, and is divided into some regions by picture portion module;Take each
The average gray in the individual region is used as the area pixel point gray value;
Greyscale transformation module, for setting I gray threshold interval Φi, by each of the image Jing after the process of described image division module
Pixel gray level replaces with each gray threshold interval Φ respectivelyiCorresponding gray scale setting valueWherein, the I is for just
Integer, the i meet 1≤i≤I;
Variation of image grayscale acquisition module, for gathering the horizontal pixel gray scale of image after the greyscale transformation resume module described in
Rise number Hup, horizontal pixel gray scale decline number Hdown;Longitudinal pixel of image after collection greyscale transformation resume module described in
Gray scale rises number Zup, longitudinal pixel grey scale decline number ZdownAnd image pixel is total after the greyscale transformation resume module described in
Number Sum;The Sum is positive integer;The Hup、Hdown、Zup、ZdownFor natural number;
Gradation of image fluctuation solves module, for obtaining the average fluctuation amplitude F of gray scale, described
The GrayjFor each pixel gray value of image after the greyscale transformation resume module described in;It is describedThe ash described in Jing
The average gray value of image after degree conversion module process;The j meets 1≤j≤Sum;
Workpiece certified products determination module, for obtaining pixel grey scale undulating value E, if E>ETH, then send sound and light alarm;It is describedThe ETHSpan is 0.01≤ETH≤10;The ETHFor pixel
Gray scale fluctuation threshold.
6. a kind of polished surface detecting system based on image procossing as claimed in claim 5, it is characterised in that described image
Grey scale change acquisition module, including:Laterally gray scale rises number solution unit, horizontal gray scale and declines number and solve unit, longitudinal direction
Gray scale rises number and solves unit and longitudinal gray scale decline number solution unit.
7. a kind of polished surface detecting system based on image procossing as claimed in claim 6, it is characterised in that it is described laterally
Gray scale rises number and solves unit, is configured to:
Gather in m rows, latter pixel gray value Gray(m,n)More than previous pixel gray value Gray(m,n-1)List
Row horizontal pixel rises number Hup_m;
Number H is risen to the single file horizontal pixel of each rowup_mSummation, obtains horizontal pixel gray scale and rises number Hup;It is described
The horizontal gray scale declines number and solves unit, is configured to:
Gather in m rows, latter pixel gray value Gray(m,n)Less than previous pixel gray value Gray(m,n-1)List
Row horizontal pixel declines number Hdown_m;
Number H is declined to the single file horizontal pixel of each rowdown_mSummation, obtains horizontal pixel gray scale and declines number Hdown;Institute
State
Wherein, the 1≤m≤M, 2≤n≤N;The M is the total line number of image, and the N is the total columns of image.
8. a kind of polished surface detecting system based on image procossing as claimed in claim 6, it is characterised in that the longitudinal direction
Gray scale rises number and solves unit, is configured to:
In the n-th row of collection, latter pixel gray value Gray(m,n)More than previous pixel gray value Gray(m-1,n)List
The longitudinal pixel of row rises number Zup_n;
Number Z is risen to described single-row longitudinal pixel of each rowup_nSummation, obtains longitudinal pixel grey scale and rises number Zup;It is described
Longitudinal gray scale declines number and solves unit, is configured to:
In the n-th row of collection, latter pixel gray value Gray(m,n)Less than previous pixel gray value Gray(m-1,n)List
The longitudinal pixel of row declines number Zdown_n;
Number Z is declined to the single file longitudinal direction pixel of each rowdown_nSummation, obtains longitudinal pixel grey scale and declines number Zdown;Institute
State
Wherein, the 2≤m≤M, 1≤n≤N;The M is the total line number of image, and the N is the total columns of image.
9. a kind of polished surface detecting system based on image procossing as claimed in claim 5, it is characterised in that:Described image
The image of acquisition module collection is MIcrosope image;The workpiece certified products determination module, also sets including gray scale fluctuation threshold
Unit;The gray scale fluctuation threshold setup unit, is configured to:Pixel grey scale fluctuation is set according to microscopical multiplication factor β
Threshold value ETH;The ETH=0.01 β;0.01≤β≤12000.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610968488.6A CN106530291B (en) | 2016-10-28 | 2016-10-28 | A kind of polished surface detection method and its system based on image procossing |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610968488.6A CN106530291B (en) | 2016-10-28 | 2016-10-28 | A kind of polished surface detection method and its system based on image procossing |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106530291A true CN106530291A (en) | 2017-03-22 |
CN106530291B CN106530291B (en) | 2019-01-22 |
Family
ID=58326659
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610968488.6A Expired - Fee Related CN106530291B (en) | 2016-10-28 | 2016-10-28 | A kind of polished surface detection method and its system based on image procossing |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106530291B (en) |
Cited By (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108287165A (en) * | 2017-11-24 | 2018-07-17 | 住华科技股份有限公司 | Defect inspection method and defect inspection system |
CN108830821A (en) * | 2018-02-08 | 2018-11-16 | 丁敏 | Household damaged degree detection method based on image procossing |
CN109030384A (en) * | 2018-07-04 | 2018-12-18 | 中国航空制造技术研究院 | A method of on-line monitoring polishing quality |
CN109514079A (en) * | 2018-12-06 | 2019-03-26 | 莆田市鑫镭腾科技有限公司 | A kind of the laser welding reinforcement means and system of the rack based on immovable fitting |
CN109754443A (en) * | 2019-01-30 | 2019-05-14 | 京东方科技集团股份有限公司 | Image data method for transformation, device and storage medium |
CN110172941A (en) * | 2019-06-14 | 2019-08-27 | 莆田市烛火信息技术有限公司 | A kind of smart city road environmental sanitation system |
CN110216530A (en) * | 2019-07-18 | 2019-09-10 | 佛山市高明金石建材有限公司 | A kind of plate grinding attachment and method for grinding |
CN110241766A (en) * | 2019-06-14 | 2019-09-17 | 莆田市烛火信息技术有限公司 | A kind of smart city road environmental sanitation method |
CN113134791A (en) * | 2021-04-30 | 2021-07-20 | 盐城工学院 | Workpiece surface treatment method based on image recognition and shot blasting machine |
WO2022142162A1 (en) * | 2020-12-29 | 2022-07-07 | 华侨大学 | Method for measuring gloss uniformity of arc-shaped surface |
CN114998343A (en) * | 2022-08-04 | 2022-09-02 | 南通广信塑料机械有限公司 | Mold surface polishing degree detection method based on vision |
CN115870815A (en) * | 2022-12-28 | 2023-03-31 | 中山超精科技有限公司 | Lens polishing detection method and device |
CN116551475A (en) * | 2023-07-04 | 2023-08-08 | 张家港市卓华金属科技有限公司 | Grinding processing method and system for hardware tool |
Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101949865A (en) * | 2010-09-19 | 2011-01-19 | 首钢总公司 | Method for optimizing Parsytec on-line surface defect detection system |
US8107716B2 (en) * | 2007-09-27 | 2012-01-31 | Fujifilm Corporation | Defect detecting apparatus and method |
CN102435628A (en) * | 2011-11-02 | 2012-05-02 | 电子科技大学 | Characterization method for structural morphology of porous electrode material |
CN102779337A (en) * | 2011-04-13 | 2012-11-14 | 南京大学 | Method for light band separation and peak positioning of structured light |
CN103196917A (en) * | 2013-03-13 | 2013-07-10 | 同济大学 | CCD linear array camera-based online rolled sheet material surface flaw detection system and detection method thereof |
US20130223673A1 (en) * | 2011-08-30 | 2013-08-29 | Digimarc Corporation | Methods and arrangements for identifying objects |
CN103885217A (en) * | 2014-03-10 | 2014-06-25 | 京东方科技集团股份有限公司 | Method and device for detecting liquid crystal display panel column-shaped shock insulator defects |
CN104568956A (en) * | 2013-10-12 | 2015-04-29 | 上海掌迪自动化科技有限公司 | Machine vision based detection method for strip steel surface defects |
CN105354831A (en) * | 2015-09-30 | 2016-02-24 | 广东工业大学 | Multi-defect detection method based on image block variance-weighting eigenvalues |
CN105938620A (en) * | 2016-04-14 | 2016-09-14 | 北京工业大学 | Small-diameter pipe inside weld surface defect identification device |
-
2016
- 2016-10-28 CN CN201610968488.6A patent/CN106530291B/en not_active Expired - Fee Related
Patent Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8107716B2 (en) * | 2007-09-27 | 2012-01-31 | Fujifilm Corporation | Defect detecting apparatus and method |
CN101949865A (en) * | 2010-09-19 | 2011-01-19 | 首钢总公司 | Method for optimizing Parsytec on-line surface defect detection system |
CN102779337A (en) * | 2011-04-13 | 2012-11-14 | 南京大学 | Method for light band separation and peak positioning of structured light |
US20130223673A1 (en) * | 2011-08-30 | 2013-08-29 | Digimarc Corporation | Methods and arrangements for identifying objects |
CN102435628A (en) * | 2011-11-02 | 2012-05-02 | 电子科技大学 | Characterization method for structural morphology of porous electrode material |
CN103196917A (en) * | 2013-03-13 | 2013-07-10 | 同济大学 | CCD linear array camera-based online rolled sheet material surface flaw detection system and detection method thereof |
CN104568956A (en) * | 2013-10-12 | 2015-04-29 | 上海掌迪自动化科技有限公司 | Machine vision based detection method for strip steel surface defects |
CN103885217A (en) * | 2014-03-10 | 2014-06-25 | 京东方科技集团股份有限公司 | Method and device for detecting liquid crystal display panel column-shaped shock insulator defects |
CN105354831A (en) * | 2015-09-30 | 2016-02-24 | 广东工业大学 | Multi-defect detection method based on image block variance-weighting eigenvalues |
CN105938620A (en) * | 2016-04-14 | 2016-09-14 | 北京工业大学 | Small-diameter pipe inside weld surface defect identification device |
Non-Patent Citations (2)
Title |
---|
DENG YAN等: "Total internal reflection microscopy:a subsurface defects identification technique in optically transparent components", 《HIGH POWER LASER AND PARTICLE BEAMS》 * |
刘健等: "光学元件亚表面损伤检测技术研究现状", 《激光与光电子学进展》 * |
Cited By (16)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108287165A (en) * | 2017-11-24 | 2018-07-17 | 住华科技股份有限公司 | Defect inspection method and defect inspection system |
CN108830821A (en) * | 2018-02-08 | 2018-11-16 | 丁敏 | Household damaged degree detection method based on image procossing |
CN109030384A (en) * | 2018-07-04 | 2018-12-18 | 中国航空制造技术研究院 | A method of on-line monitoring polishing quality |
CN109514079A (en) * | 2018-12-06 | 2019-03-26 | 莆田市鑫镭腾科技有限公司 | A kind of the laser welding reinforcement means and system of the rack based on immovable fitting |
CN109754443A (en) * | 2019-01-30 | 2019-05-14 | 京东方科技集团股份有限公司 | Image data method for transformation, device and storage medium |
CN110172941B (en) * | 2019-06-14 | 2021-05-04 | 广东特丽洁环境工程有限公司 | Wisdom urban road sanitation system |
CN110241766A (en) * | 2019-06-14 | 2019-09-17 | 莆田市烛火信息技术有限公司 | A kind of smart city road environmental sanitation method |
CN110172941A (en) * | 2019-06-14 | 2019-08-27 | 莆田市烛火信息技术有限公司 | A kind of smart city road environmental sanitation system |
CN110216530A (en) * | 2019-07-18 | 2019-09-10 | 佛山市高明金石建材有限公司 | A kind of plate grinding attachment and method for grinding |
WO2022142162A1 (en) * | 2020-12-29 | 2022-07-07 | 华侨大学 | Method for measuring gloss uniformity of arc-shaped surface |
CN113134791A (en) * | 2021-04-30 | 2021-07-20 | 盐城工学院 | Workpiece surface treatment method based on image recognition and shot blasting machine |
CN114998343A (en) * | 2022-08-04 | 2022-09-02 | 南通广信塑料机械有限公司 | Mold surface polishing degree detection method based on vision |
CN115870815A (en) * | 2022-12-28 | 2023-03-31 | 中山超精科技有限公司 | Lens polishing detection method and device |
CN115870815B (en) * | 2022-12-28 | 2023-06-16 | 中山超精科技有限公司 | Lens polishing detection method and device |
CN116551475A (en) * | 2023-07-04 | 2023-08-08 | 张家港市卓华金属科技有限公司 | Grinding processing method and system for hardware tool |
CN116551475B (en) * | 2023-07-04 | 2023-09-22 | 张家港市卓华金属科技有限公司 | Grinding processing method and system for hardware tool |
Also Published As
Publication number | Publication date |
---|---|
CN106530291B (en) | 2019-01-22 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106530291A (en) | Polished surface detection method and system based on image processing | |
CN104794491B (en) | Based on the fuzzy clustering Surface Defects in Steel Plate detection method presorted | |
CN106471333B (en) | Flaw detection apparatus and production system | |
CN107845090B (en) | Silicon wafer detection method and silicon wafer detection device | |
Çelik et al. | Development of a machine vision system: real-time fabric defect detection and classification with neural networks | |
CN107157447B (en) | Skin surface roughness detection method based on image RGB color space | |
CN107478657A (en) | Stainless steel surfaces defect inspection method based on machine vision | |
CN106546197B (en) | Based on horizontal self-alignment polished surface detection method and its system | |
CN101619964A (en) | Quick microscopic detection method and quick microscopic detection device of micro accessory based on process match | |
CN101634551A (en) | Method and system for detecting surface roughness | |
CN109974648B (en) | Method for evaluating geometric accuracy of micro-hole based on coordinate value | |
CN104568956B (en) | The detection method of the steel strip surface defect based on machine vision | |
CN114219794B (en) | Method and system for evaluating surface quality of shaving board based on machine vision | |
CN106770321A (en) | A kind of plastic part defect inspection method interval based on multi thresholds | |
CN110766683B (en) | Pearl finish grade detection method and system | |
CN115063620A (en) | Bit layering-based Roots blower bearing wear detection method | |
CN105300302A (en) | Diameter measurement method for Brinell hardness indent circle | |
CN114998308A (en) | Defect detection method and system based on photometric stereo | |
CN116008289A (en) | Nonwoven product surface defect detection method and system | |
CN113034482A (en) | Surface roughness detection method based on machine vision and machine learning | |
CN110873718A (en) | Steel plate surface defect detection system and method based on machine vision | |
Chen et al. | A study of a rapid method for detecting the machined surface roughness | |
CN114881998A (en) | Workpiece surface defect detection method and system based on deep learning | |
CN114881969A (en) | Wood knot detection method and detection system based on computer vision on wood board surface | |
CN104515473A (en) | Online diameter detection method of varnished wires |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
CF01 | Termination of patent right due to non-payment of annual fee | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20190122 Termination date: 20191028 |