CN110687121B - Intelligent online detection and automatic grading method and system for ceramic tiles - Google Patents
Intelligent online detection and automatic grading method and system for ceramic tiles Download PDFInfo
- Publication number
- CN110687121B CN110687121B CN201910888799.5A CN201910888799A CN110687121B CN 110687121 B CN110687121 B CN 110687121B CN 201910888799 A CN201910888799 A CN 201910888799A CN 110687121 B CN110687121 B CN 110687121B
- Authority
- CN
- China
- Prior art keywords
- image
- defect
- detection
- image data
- ceramic tiles
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 239000000919 ceramic Substances 0.000 title claims abstract description 104
- 238000001514 detection method Methods 0.000 title claims abstract description 102
- 238000000034 method Methods 0.000 title claims abstract description 35
- 230000007547 defect Effects 0.000 claims abstract description 120
- 238000001914 filtration Methods 0.000 claims abstract description 28
- 238000000605 extraction Methods 0.000 claims abstract description 15
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 14
- 239000003086 colorant Substances 0.000 claims abstract description 5
- 238000003709 image segmentation Methods 0.000 claims abstract description 5
- 238000007781 pre-processing Methods 0.000 claims description 26
- 238000012545 processing Methods 0.000 claims description 16
- 230000011218 segmentation Effects 0.000 claims description 15
- 230000002950 deficient Effects 0.000 claims description 12
- 238000004364 calculation method Methods 0.000 claims description 8
- 239000000284 extract Substances 0.000 claims description 6
- 238000005286 illumination Methods 0.000 claims description 6
- 210000003298 dental enamel Anatomy 0.000 claims description 5
- 230000007797 corrosion Effects 0.000 claims description 4
- 238000005260 corrosion Methods 0.000 claims description 4
- 230000003287 optical effect Effects 0.000 claims description 4
- 230000009466 transformation Effects 0.000 claims description 3
- 230000002159 abnormal effect Effects 0.000 claims description 2
- 230000007812 deficiency Effects 0.000 abstract description 3
- 230000006872 improvement Effects 0.000 description 10
- 239000002699 waste material Substances 0.000 description 8
- 238000012360 testing method Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 238000004519 manufacturing process Methods 0.000 description 4
- 239000011159 matrix material Substances 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 229910000831 Steel Inorganic materials 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000877 morphologic effect Effects 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000012797 qualification Methods 0.000 description 1
- 239000010959 steel Substances 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
- 230000004382 visual function Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/20—Image enhancement or restoration using local operators
- G06T5/30—Erosion or dilatation, e.g. thinning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
- G01N2021/8854—Grading and classifying of flaws
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
- G01N2021/8887—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges based on image processing techniques
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Signal Processing (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The invention discloses a method and a system for intelligent online detection and automatic grading of ceramic tiles, which are used for respectively carrying out crack defect detection, unfilled corner defect detection, color difference defect detection, glaze deficiency defect detection and bulge defect detection on image data by acquiring the image data of ceramic tiles to be detected, acquiring the grade of the ceramic tiles to be detected according to the defect detection result, respectively carrying out first pretreatment or second pretreatment on the acquired image according to the detection requirement, simultaneously adopting different filtering and defect extraction modes, particularly carrying out color discrimination during the glaze deficiency detection, and adopting different algorithms to carry out image pretreatment, image segmentation, feature extraction and corresponding defect grade on the ceramic tiles to be detected with different colors, thereby realizing multiple defect detection and realizing intelligent grading of the ceramic tiles with multiple complex appearance structures.
Description
Technical Field
The invention belongs to the field of image detection, and particularly relates to an intelligent online detection and automatic grading method and system for ceramic tiles.
Background
The quality detection of the ceramic tile is an important and indispensable link in the production process of the ceramic tile, and at present, many ceramic tile production enterprises still adopt a manual detection method to complete quality detection and product classification, so that the problems of large detection error and high omission factor are solved.
Some organizations have developed researches on quality detection of ceramic products by using machine vision technology at home and abroad. The machine vision technology is a cross discipline in many fields such as artificial intelligence, neurobiology, psychophysics, computer science, image processing and pattern recognition. Machine vision mainly uses a computer to simulate the visual function of a human, extracts information from an image of an objective object, processes and understands the information, and finally is used for actual detection, measurement and control. The machine vision technology has the biggest characteristics of high speed, large information amount and multiple functions.
However, most of the existing researches are carried out on certain single defects such as cracks, glaze shortage, chromatic aberration and the like, and the system and the intelligence are not enough. Secondly, the existing research mainly focuses on ceramic products such as ceramic tiles, ceramic tiles and ceramic bowls with simple geometric shapes, and the defects of the ceramic tile products with complicated geometric shapes can not be detected simultaneously, and the intelligent classification of the ceramic tile detection with various complicated shape structures can not be realized.
Disclosure of Invention
Aiming at the defects or improvement requirements in the prior art, the invention provides an intelligent online detection and automatic grading method and system for ceramic tiles, which are used for acquiring image data of ceramic tiles to be detected, respectively performing crack defect detection, unfilled corner defect detection, color difference defect detection, glaze defect detection and bulge defect detection on the image data, acquiring the grade of the ceramic tiles to be detected according to the defect detection result, respectively performing first preprocessing or second preprocessing on the acquired image according to the detection requirement, and meanwhile, adopting different filtering and defect extraction modes, so that various defect detections can be realized, and intelligent grading of the ceramic tiles with various complex appearance structures can be realized.
In order to achieve the above object, according to one aspect of the present invention, there is provided an intelligent online detection and automatic classification method for ceramic tiles, comprising the steps of:
acquiring image data of the ceramic tile to be detected, respectively carrying out crack defect detection, unfilled corner defect detection, color difference defect detection, enamel defect detection and bulge defect detection on the image data, acquiring the grade of the ceramic tile to be detected according to the defect detection result, wherein,
the crack defect detection is to perform first preprocessing on image data, extract defect characteristics by adopting a user-defined sliding filtering and automatic region growing method, and judge whether the ceramic tile to be detected has crack defects and corresponding defect grades according to the extracted defect characteristics;
the unfilled corner defect detection comprises the steps of carrying out second preprocessing on image data, extracting unfilled corner features of the ceramic tile by adopting a self-adaptive threshold segmentation algorithm and calculating the unfilled corner area;
performing second preprocessing on the image data, extracting the pottery tile color difference characteristic by adopting a self-adaptive threshold segmentation algorithm and calculating the color difference area;
the glaze-lacking defect detection is to judge the color of the ceramic tile to be detected according to the image data, and carry out image preprocessing, image segmentation, feature extraction and corresponding defect grade on the ceramic tile to be detected with different colors by adopting different algorithms;
and the bulge defect detection is to carry out second pretreatment on the image data, filter the pretreated image by respectively adopting a user-defined sliding filter, linear median filter and interpolation low-pass filter, extract defect characteristics by adopting a threshold segmentation method, and judge whether the ceramic tile to be detected has a crack defect and a corresponding defect grade according to the extracted defect characteristics.
As a further improvement of the invention, the first preprocessing is to perform color space transformation on the image data, adopt the image of the red channel in the RGB image, and select a preset template to perform median filtering on the image of the red channel.
As a further improvement of the invention, the self-defined sliding filtering is to adopt a double-window model to perform sliding filtering on the image after the first preprocessing, wherein an outer window of the double-window model is a background area, and an inner window is a detection area.
As a further improvement of the method, the automatic region growing method specifically comprises the steps of carrying out local window scanning on a crack region, and determining a seed point by comparing the numerical values of the mean value and the central value of a local window; setting a threshold value for stopping growth, calculating the absolute value of the pixel difference between the point to be marked and the adjacent point, taking the point to be marked as a seed point, continuing the regional growth of the point to be marked, and traversing all the points to be marked to extract the seed point.
As a further improvement of the present invention, the second preprocessing is graying the image data.
As a further improvement of the present invention, the color discrimination specifically comprises:
and converting the image data into an image in an HSV format, and comparing the average values of the three channels to judge the color of the to-be-detected ceramic tile.
As a further improvement of the method, the pottery tile to be detected is of a first color, image data is converted into an image of an NTSV format, Gaussian filtering is respectively carried out on three channels of the image of the NTSV format, then global threshold segmentation is respectively carried out on the two-channel image and the three-channel image to obtain a binary image of the image to be detected, and corrosion operation is carried out on the binary image to realize defect feature extraction;
and converting the image data into an image in an HSV format when the to-be-detected ceramic tile is in the second color, and segmenting the image in the HSV format by adopting a light and shade anomaly detection algorithm so as to realize defect feature extraction.
In order to achieve the above object, according to another aspect of the present invention, an intelligent online detection and automatic classification system for ceramic tiles is provided, which includes an image acquisition system and an image processing system, wherein the image acquisition system is configured to acquire an image of a ceramic tile to be detected, and the image processing system is configured to receive acquired image data of the image acquisition system and implement the above method.
As a further improvement of the invention, the image acquisition system comprises a distributed LED shadow-free illumination system, a photoelectric switch, an area array CMOS camera and an optical lens.
As a further improvement of the invention, the distributed LED shadowless illumination system comprises a dark box and a light source, wherein the light source comprises a plurality of LED lamp belts which are arranged in parallel at the top end inside the dark box.
Generally, compared with the prior art, the above technical solution conceived by the present invention has the following beneficial effects:
according to the intelligent online detection and automatic grading method and system for the ceramic tiles, the collected ceramic tile images are sequentially subjected to line crack defect detection, unfilled corner defect detection, color difference defect detection, glaze deficiency defect detection and bulge defect detection, the grade of the ceramic tiles to be detected is obtained according to the defect detection results, and the ceramic tiles with various types of defects can be screened out. Compared with the traditional manual detection, the technical scheme of the invention adopts a machine vision method, takes pictures of the ceramic tiles on the production line by utilizing the color area array CCD to obtain the surface images of the ceramic tiles, completes the detection and identification of the surface defects by the image processing technology, and finally automatically grades the ceramic tiles according to the detection result, so that all the ceramic tiles can be rapidly and efficiently detected one by one, the classification effect is good, the production efficiency can be greatly improved, and the qualification rate of the ceramic tiles leaving the factory is obviously improved.
According to the intelligent online detection and automatic grading method and system for the ceramic tiles, the collected images are respectively subjected to first preprocessing and second preprocessing according to detection requirements, meanwhile, different filtering and defect extraction modes are adopted, particularly, color discrimination is carried out during glaze shortage detection, different algorithms are adopted to carry out image preprocessing, image segmentation, feature extraction and corresponding defect grades on the ceramic tiles to be detected in different colors, and therefore multiple defect detections can be achieved, and intelligent grading of the ceramic tiles with multiple complex appearance structures is achieved.
Drawings
FIG. 1 is a schematic diagram of an intelligent online detection and automatic classification method for ceramic tiles according to an embodiment of the present invention;
fig. 2 is a schematic structural diagram of an intelligent online detection and automatic classification system for ceramic tiles according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and do not limit the invention.
In addition, the technical features involved in the embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other. The present invention will be described in further detail with reference to specific embodiments.
Fig. 1 is a schematic diagram of an intelligent online detection and automatic classification method for a ceramic tile according to an embodiment of the present invention. As shown in fig. 1, an intelligent online detection and automatic classification method for ceramic tiles includes the following steps:
acquiring image data of the ceramic tile to be detected, respectively carrying out crack defect detection, unfilled corner defect detection, color difference defect detection, enamel defect detection and bulge defect detection on the image data, acquiring the grade of the ceramic tile to be detected according to the defect detection result, wherein,
the crack defect detection is to perform first preprocessing on image data, extract defect characteristics by adopting a user-defined sliding filtering and automatic region growing method, and judge whether the ceramic tile to be detected has crack defects and corresponding defect grades according to the extracted defect characteristics;
the first preprocessing specifically comprises the steps of carrying out color space transformation on image data, adopting an image of a red channel in an RGB image, selecting a preset template to carry out median filtering on the image of the red channel so as to filter noise points in the image data, increase the contrast and facilitate crack extraction;
specifically, the self-defined sliding filtering is to perform sliding filtering on the first preprocessed image by adopting a double-window model, wherein an outer window of the double-window model is a background area, an inner window of the double-window model is a detection area, the contrast of crack defects can be increased, and a crack area is extracted by adopting global threshold segmentation;
the automatic region growing method specifically comprises the steps of carrying out local window scanning on a crack region, and determining a seed point by comparing the numerical values of the mean value and the central value of a local window; setting a threshold value for stopping growth, calculating the absolute value of the pixel difference between the point to be marked and the adjacent point, taking the point to be marked as a seed point, continuing the regional growth of the point to be marked, otherwise, stopping the growth, traversing all the points to be marked until the growth of the seed point is finished, and extracting all the seed points.
Calculating the length-width ratio and the area of the extracted defect, and judging whether the defect exists, wherein the calculation formula of the defect area is as follows:
wherein f (x, y) is a defective pixel point, I is a defective region, A is a defective total pixel point,
the aspect ratio is calculated as:
wherein, NxIs the longest length of the defect, NyIs the maximum width of the defect region, and B is the aspect ratio.
When A is more than 50 pixels and B is more than 5, it can be judged that the defect is a crack defect.
The unfilled corner defect detection comprises the steps of carrying out second preprocessing on image data, extracting unfilled corner features of the ceramic tile by adopting a self-adaptive threshold segmentation algorithm and calculating the unfilled corner area;
the second preprocessing is to carry out graying on the image data;
performing second preprocessing on the image data, extracting the pottery tile color difference characteristic by adopting a self-adaptive threshold segmentation algorithm and calculating the color difference area;
the glaze-lacking defect detection is to judge the color of the ceramic tile to be detected according to the image data, and carry out image preprocessing, image segmentation, feature extraction and corresponding defect grade on the ceramic tile to be detected with different colors by adopting different algorithms;
the specific color discrimination of the ceramic tile to be detected according to the image data is as follows:
converting the image data into an HSV (hue, saturation and value) format image, and comparing the average values of the three channels to judge the color of the to-be-detected ceramic tile;
the method comprises the steps that a to-be-detected ceramic tile is in a first color, the first color is red as an example, image data are converted into an image in an NTSV format, Gaussian filtering is conducted on three channels of the image in the NTSV format respectively, then global threshold segmentation is conducted on the two-channel image and the three-channel image respectively to obtain a binary image of the to-be-detected image, corrosion operation is conducted on the binary image to achieve defect feature extraction, logical and operation are conducted on the two-channel image and the three-channel image to remove interference points in a background area in the three channels, defects in the three channels are obviously distinguished from background portions, and a final binary image is obtained;
the pottery tile to be detected is of a second color, the second color is blue as an example, the image data is converted into an image in an HSV (hue, saturation and value) format, and the image in the HSV format is segmented by adopting a light and shade anomaly detection algorithm so as to realize defect feature extraction;
specifically, after the image is normalized, the pottery tile area is divided at equal intervals, the mean value and the variance of the divided area are calculated, f (x, y) is a pixel value, wherein A multiplied by B represents the size of the divided area.
d=μ-3σ
Under the assumption of normal distribution, the area mu +/-3 sigma contains 99.7% of data, if the mean value mu of a certain value distance distribution exceeds 3 sigma, if f (x, y) is smaller than f, the pixel point is an abnormal point;
judging the corresponding glaze-lacking defect level, namely, extracting the length characteristic of each connected domain of the binary image;
analyzing and judging the characteristic value to obtain whether the ceramic tile has the defect of glaze shortage on the surface;
let NxIs the sum of the pixels in the x direction, NyIs a y directionUp pixel sum, the length of the defect L is:
and L is less than 30 pixels, judging that the glaze shortage does not exist, otherwise, outputting a glaze shortage parameter.
f (x, y) is a defective pixel value, and the area S of the enamel defect is as follows:
the bulge defect detection is to carry out second preprocessing on image data, filter the preprocessed image by respectively adopting a user-defined sliding filter, linear median filtering and interpolation low-pass filtering so as to improve the contrast ratio of a bulge region and a background region on the surface of the ceramic tile, extract defect characteristics by adopting a threshold segmentation method, and judge whether the ceramic tile to be detected has crack defects and corresponding defect grades according to the extracted defect characteristics;
the method for extracting the defects by adopting the threshold segmentation method specifically comprises the following steps of selecting a pottery tile region to be detected in a preprocessed image: comparing the row and column mean values of the preprocessed gray level image with the total mean value to obtain the to-be-detected ceramic tile area W1The calculation formula is as follows:
wherein, the size of the original image gray level image is mxn; w0(i, j) is a grayscale image W0Image of (1)The prime value.
Expanding the left edge and the right edge of the selected area to be detected: setting the area of the ceramic tile to be detected as m1×n1Gray scale matrix W1Then from W1Row mean value pair of (W)1And expanding left and right. Let W1Has a line mean value of K1=[K1(1);...;K1(l)](l=1,2,...,m1) Let D ═ 1,. 1]D is 1 XN; and L is equal to K1D, its size is m1And (4) times N. Let the extended matrix be W2I.e. W2=[L,W1,L]Having a size of m1×(n1+2N);
A filter H is constructed with a size of 1 × N. The formula is as follows:
H=[H1,H2,H1]=[h(1),...,h(N)]
wherein H1Is 1 XN1,H2Is 1 XN2And has 2N1+N2=N
To W2Performing sliding filter processing in horizontal direction to generate W3: let W2=[X(1);...;X(m1)](ii) a Wherein W3=[Y(1);...;Y(m1)], The calculation formula is as follows:
yk(j)=h(1)xk(j)+h(2)xk-1(j)+…+h(N)xk-N+1(j)
(k=1,...,n1+2N;j=1,...,m1)
then to W3Performing extension of upper and lower edges to obtain W4: from W3Column average of
S=[S(1),...,S(r)](r=1,2,...,n1+2N)
To obtain L1=DTS, its size is Nx (N)1+ 2N). Then W is4=[L1;W3;L1]Having a size of (m)1+2N)×(n1+2N)。
Using linear filters HTTo W4Performing sliding filter processing in vertical direction to generate matrix W5: let W4=[A(1),...,A(n1+2N)],
W5=[B(1),...,B(n1+2N)]
then the calculation formula is:
bk(j)=h(1)ak(j)+h(2)ak-1(j)+…+h(N)ak-N+1(j)
(k=1,...,m1+2N;j=1,...,n1+2N)
to W5Carrying out one-dimensional median filtering processing in the horizontal direction to obtain W6(ii) a Then from W5And W6Subtracting to obtain W7(ii) a And then to W7And selecting the area, wherein the calculation formula is as follows:
W7=W5-W6
wherein W5,W6,W7Has a size of (m)1+2N)×(n1+2N)。
To W7To carry outRegion selection to obtain W8,W8Is m in size1×n1The calculation formula is as follows:
W8=W7(i,j)
The contrast ratio of the pottery tile bulge area and the background area is greatly increased through the operation, and the gray threshold value T is utilized1To W8And carrying out binarization to realize the preliminary segmentation of the bulges. Calculating the formula:
obtaining a corresponding binary image, namely:
i=1,...,m1;j=1,...,n1
and (3) secondary division of the bulge area: from W9And judging whether bulges exist on the surface of the ceramic tile. If there is a bulge, return to W3Bulge region is obtained as0To A, a0By interpolating low-pass filter FkAnd performing R times of filtering processing, wherein the size of the image before and after each filtering processing is the same. The calculation formula is as follows:
the interpolation low-pass filter corresponding to the k filtering is as follows:
Fk=Fk2 k=1,2,...,R
wherein: q ═ 1, 2, (k +1) × j-k; 1, 2, (k +1) × i-k; 1, 2, ·, 5; j ═ 1, 2,. 5; k 1, 2.
To the bulge area A0Filtering for the kth time to obtain:
A{0}=A0
A{k}=A{k-1}-A{k-1}*Fk k=1,2,...,R
from the filtered image a { k }, k ═ 1, 2
For B may be defined by a threshold value T2Performing binarization treatment to obtain B1Wherein the threshold value is T2Take 0.06 and filter times R15. And finally, obtaining the bulge area of the steel pipe through morphological corrosion and expansion treatment.
Fig. 2 is a schematic structural diagram of an intelligent online detection and automatic classification system for ceramic tiles according to an embodiment of the present invention. As shown in fig. 2, an intelligent online detection and automatic classification system for ceramic tiles comprises an image acquisition system and an image processing system, wherein the image acquisition system is used for acquiring images of ceramic tiles to be detected, and the image processing system is used for receiving acquired image data of the image acquisition system and implementing the method;
the image acquisition system comprises a distributed LED shadow-free illumination system, a photoelectric switch, an area array CMOS camera and an optical lens;
the distributed LED shadowless illumination system consists of a dark box and a light source. The camera bellows is a box with a three-dimensional structure, and provides a camera chamber environment for image acquisition; the light source is composed of a plurality of LED lamp belts which are arranged in parallel at the top end in the dark box, so that the ceramic tile curved surface shadow and enamel reflection are eliminated, uniform illumination is provided for the image acquisition system, and the requirement of high-quality image acquisition is met.
The image acquisition system also comprises a photoelectric switch, an area array CMOS camera and a corresponding optical lens. And a high-definition and high-quality ceramic tile image is obtained, so that the rapid and accurate processing of an image processing system is facilitated. And when the photoelectric switch detects the ceramic tile, the CMOS camera is triggered to automatically photograph the front surface of the ceramic tile.
The image processing system can be realized by an industrial personal computer, and accurate grading is realized by combining software according to a grading standard and a grading database. After the CMOS camera automatically shoots the front surface of the ceramic tile, the shot image is transmitted to the industrial personal computer through the gigabit Ethernet port for grading detection, and the industrial personal computer transmits grading results through 4 paths of signals through the I/O port (1 path of signal corresponds to 1 grade of the ceramic tile).
As an example, the classification results of the detection of the ceramic tile defect are respectively: superior, first-class, qualified, waste;
a waste product comprising: if the length and the width of the crack are larger than the maximum value of the length and the width of the crack of the preset ceramic tile, the crack is a waste product; bulging, and taking the bulge as a waste product if the bulge area is larger than the maximum value of the bulge area of the preset ceramic tile; if the glaze-lacking area is larger than the preset maximum value of the glaze-lacking area of the ceramic tile, the ceramic tile is a waste product; if the color difference area is larger than the maximum value of the preset color difference area of the ceramic tile, the ceramic tile is taken as a waste product; and (4) a defective corner, namely, if the ratio of the defective corner area to the whole ceramic tile is greater than the preset maximum ratio of the defective corner area to the whole ceramic tile, the defective corner is taken as a waste product.
In order to verify the performance of the system, a series of test experiments are developed, including a detection speed test, a missing detection rate test, a grading accuracy rate test and the like. The test results were as follows:
(1) the detection speed of the ceramic tile is less than 1.05 s/sheet;
(2) the omission rate of finished products is less than 3 percent;
(3) the grading accuracy is more than 95 percent;
(4) can realize the automatic classification of superior products, first-class products, qualified products and waste products.
It will be understood by those skilled in the art that the foregoing is only a preferred embodiment of the present invention, and is not intended to limit the invention, and that any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the scope of the present invention.
Claims (8)
1. An intelligent online detection and automatic grading method for ceramic tiles is characterized by comprising the following steps:
acquiring image data of the ceramic tile to be detected, respectively carrying out crack defect detection, unfilled corner defect detection, color difference defect detection, enamel defect detection and bulge defect detection on the image data, and acquiring the grade of the ceramic tile to be detected according to the defect detection result, wherein,
the crack defect detection comprises the steps of carrying out first preprocessing on image data, extracting defect characteristics by adopting a user-defined sliding filtering and automatic region growing method, and judging whether crack defects and corresponding defect grades exist in the ceramic tiles to be detected according to the extracted defect characteristics;
calculating the length-width ratio and the area of the extracted defect characteristics, and judging whether the defect characteristics have defects or not, wherein the calculation formula of the defect area is as follows:
wherein f (x, y) is a defective pixel, I is a defective region, A is a defective total pixel,
the aspect ratio is calculated as:
wherein N isxIs the longest length of the defective area, NyIs the maximum width of the defect region, and B is the aspect ratio;
when A is larger than 50 pixels and B is larger than 5, judging the defect as a crack defect;
the unfilled corner defect detection comprises the steps of carrying out second preprocessing on image data, extracting unfilled corner features of the ceramic tile by adopting a self-adaptive threshold segmentation algorithm and calculating the unfilled corner area;
the color difference defect detection is to perform second preprocessing on the image data, extract the pottery tile color difference characteristics by adopting a self-adaptive threshold segmentation algorithm and calculate the color difference area;
the glaze-lacking defect detection is to judge the color of the ceramic tile to be detected according to image data, and carry out image preprocessing, image segmentation, feature extraction and corresponding defect grade on the ceramic tile to be detected with different colors by adopting different algorithms;
the color discrimination specifically comprises:
converting the image data into an HSV (hue, saturation and value) format image, and comparing the average values of the three channels to judge the color of the to-be-detected ceramic tile;
converting image data into an image in an NTSV format, performing Gaussian filtering on three channels of the image in the NTSV format respectively, performing global threshold segmentation on the two-channel image and the three-channel image respectively to obtain a binary image of the image to be detected, and performing corrosion operation on the binary image to realize defect feature extraction;
converting the image data into an image in an HSV format when the to-be-detected ceramic tile is in a second color, and segmenting the image in the HSV format by adopting a light and shade anomaly detection algorithm to realize defect feature extraction;
firstly, normalizing an image, then, segmenting a ceramic tile region at equal intervals, computing the mean value and the variance of the segmented region, wherein f (x, y) is a pixel value, and A multiplied by B represents the size of the segmented region;
d=μ-3σ
if the mean value mu of the distance distribution of a certain value exceeds 3 sigma, if f (x, y) is smaller than f, the pixel point is an abnormal point;
judging the corresponding glaze-lacking defect grade, namely, extracting the length characteristic of each connected domain of the binary image;
analyzing and judging the characteristic value to obtain whether the ceramic tile has the defect of glaze shortage on the surface;
let NxIs the sum of the pixels in the x direction, NyThe length of the defect, L, is the sum of the pixels in the y-direction:
if L is less than 30 pixels, judging that no glaze shortage exists, otherwise, outputting a glaze shortage parameter;
and the bulge defect detection comprises the steps of carrying out second preprocessing on image data, filtering the preprocessed image by respectively adopting a user-defined sliding filter, linear median filtering and interpolation low-pass filtering, extracting defect characteristics by adopting a threshold segmentation method, and judging whether the ceramic tile to be detected has a crack defect and a corresponding defect grade according to the extracted defect characteristics.
2. The intelligent online detection and automatic classification method for the ceramic tiles as claimed in claim 1, wherein the first preprocessing is to perform color space transformation on image data, adopt an image of a red channel in an RGB image, and select a preset template to perform median filtering on the image of the red channel.
3. The intelligent online detection and automatic classification method for the ceramic tiles according to claim 1 or 2, wherein the customized sliding filtering is sliding filtering of the first preprocessed image by using a double-window model, an outer window of the double-window model is a background area, and an inner window of the double-window model is a detection area.
4. The intelligent online detection and automatic classification method for the ceramic tiles according to claim 3, wherein the automatic region growing method is specifically characterized in that local window scanning is carried out on a crack region, and a seed point is determined by comparing the values of the mean value and the central value of the local window; setting a threshold value for stopping growth, calculating the absolute value of the pixel difference between the point to be marked and the adjacent point, taking the point to be marked as a seed point, continuing the regional growth of the point to be marked, and traversing all the points to be marked to extract the seed point.
5. The intelligent online detection and automatic classification method for ceramic tiles as claimed in claim 1, wherein the second preprocessing is graying image data.
6. An intelligent online detection and automatic grading system for ceramic tiles, which comprises an image acquisition system and an image processing system, and is characterized in that the image acquisition system is used for acquiring images of ceramic tiles to be detected, and the image processing system is used for receiving acquired image data of the image acquisition system and realizing the method of any one of claims 1 to 5.
7. The intelligent online detection and automatic grading system for ceramic tiles according to claim 6, wherein the image acquisition system comprises a distributed LED shadowless illumination system, a photoelectric switch, an area-array CMOS camera and an optical lens.
8. The intelligent online detection and automatic grading system for ceramic tiles as claimed in claim 7, wherein the distributed LED shadowless lighting system comprises a dark box and a light source, and the light source comprises a plurality of LED strips arranged in parallel at the top end inside the dark box.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910888799.5A CN110687121B (en) | 2019-09-19 | 2019-09-19 | Intelligent online detection and automatic grading method and system for ceramic tiles |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910888799.5A CN110687121B (en) | 2019-09-19 | 2019-09-19 | Intelligent online detection and automatic grading method and system for ceramic tiles |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110687121A CN110687121A (en) | 2020-01-14 |
CN110687121B true CN110687121B (en) | 2022-06-17 |
Family
ID=69109641
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910888799.5A Active CN110687121B (en) | 2019-09-19 | 2019-09-19 | Intelligent online detection and automatic grading method and system for ceramic tiles |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110687121B (en) |
Families Citing this family (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111558542A (en) * | 2020-06-24 | 2020-08-21 | 重庆视智迪科技有限公司 | Ceramic tile surface quality online detection sorting system and method |
CN112884740A (en) * | 2021-02-20 | 2021-06-01 | 联想(北京)有限公司 | Image detection method and device, electronic equipment and storage medium |
CN113405950B (en) * | 2021-07-22 | 2022-07-05 | 福建恒安集团有限公司 | Method for measuring diffusion degree of disposable sanitary product |
CN113744263B (en) * | 2021-09-17 | 2023-10-27 | 景德镇陶瓷大学 | Method for rapidly detecting surface defects of small-size mosaic ceramic |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104101600A (en) * | 2013-04-15 | 2014-10-15 | 宝山钢铁股份有限公司 | Method and apparatus for detecting fine cracks on cross section of continuous casting slab |
Family Cites Families (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103440674B (en) * | 2013-06-13 | 2016-06-22 | 厦门美图网科技有限公司 | A kind of rapid generation of digital picture wax crayon specially good effect |
CN103872983B (en) * | 2014-04-04 | 2016-05-25 | 天津市鑫鼎源科技发展有限公司 | Solar cell surface defect inspection method |
CN106204614B (en) * | 2016-07-21 | 2019-01-08 | 湘潭大学 | A kind of workpiece appearance defects detection method based on machine vision |
CN106353324B (en) * | 2016-08-10 | 2019-01-18 | 浙江理工大学 | Magnet ring method for extracting surface defects |
CN107871302A (en) * | 2016-09-23 | 2018-04-03 | 电子科技大学 | A kind of Infrared Image Pseudo-Color processing method based on YUV color spaces |
CN108694711A (en) * | 2018-02-19 | 2018-10-23 | 江苏新时高温材料股份有限公司 | Ceramic membrane surface defect two-dimensional detecting method is realized based on machine vision technique |
CN108445009B (en) * | 2018-03-06 | 2021-03-16 | 大连海事大学 | Crack detection method for solar cell panel |
CN109523505B (en) * | 2018-09-18 | 2021-06-11 | 深圳市智信精密仪器有限公司 | Method for detecting pattern defects on surface of ceramic tile based on machine vision |
CN109727244B (en) * | 2019-01-18 | 2020-12-08 | 深圳至汉装备科技有限公司 | Magnetic shoe surface crack detection method |
-
2019
- 2019-09-19 CN CN201910888799.5A patent/CN110687121B/en active Active
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104101600A (en) * | 2013-04-15 | 2014-10-15 | 宝山钢铁股份有限公司 | Method and apparatus for detecting fine cracks on cross section of continuous casting slab |
Non-Patent Citations (2)
Title |
---|
基于非下采样Shearlet变换的磁瓦表面裂纹检测;杨成立等;《农业机械学报》;20170331;第48卷(第03期);第405-412页 * |
磁瓦表面缺陷自动检测***的研究;严俊龙等;《计算机工程与应用》;20091231;第45卷(第36期);第228-231页 * |
Also Published As
Publication number | Publication date |
---|---|
CN110687121A (en) | 2020-01-14 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110687121B (en) | Intelligent online detection and automatic grading method and system for ceramic tiles | |
CN109269951B (en) | Image-based flotation tailing ash content, concentration and coarse particle content detection method | |
CN110415181B (en) | Intelligent identification and grade judgment method for RGB (red, green and blue) images of flue-cured tobacco in open environment | |
CN109872300B (en) | Visual saliency detection method for appearance defects of friction plate | |
CN104715239B (en) | A kind of vehicle color identification method based on defogging processing and weight piecemeal | |
CN109191459A (en) | The automatic identification and ranking method of continuous casting billet macrostructure center segregation defect | |
CN108765402B (en) | Non-woven fabric defect detection and classification method | |
CN104537651B (en) | Proportion detecting method and system for cracks in road surface image | |
CN112149543B (en) | Building dust recognition system and method based on computer vision | |
CN111667475B (en) | Machine vision-based Chinese date grading detection method | |
CN108197521A (en) | A kind of leggy Quick Response Code obtains identification device and method | |
CN110807763A (en) | Method and system for detecting ceramic tile surface bulge | |
CN110111322A (en) | A kind of tunnel defect identifying system based on image | |
CN115082451A (en) | Stainless steel soup ladle defect detection method based on image processing | |
CN110687122A (en) | Method and system for detecting surface cracks of ceramic tile | |
CN115100206A (en) | Printing defect identification method for textile with periodic pattern | |
CN112561875A (en) | Photovoltaic cell panel coarse grid detection method based on artificial intelligence | |
CN111985435A (en) | Unmanned aerial vehicle water area monitoring and cruising method based on machine vision | |
CN111738931A (en) | Shadow removal algorithm for aerial image of photovoltaic array unmanned aerial vehicle | |
CN110443166A (en) | A kind of licence plate recognition method of haze weather | |
CN111985436A (en) | Workshop ground mark line identification fitting method based on LSD | |
CN107239761A (en) | Fruit tree branch pulling effect evaluation method based on skeleton Corner Detection | |
CN111192332B (en) | Smoke detection-based smoke machine control method and smoke machine | |
CN108830834A (en) | A kind of cable-climbing robot video artefacts information automation extraction method | |
CN113744263B (en) | Method for rapidly detecting surface defects of small-size mosaic ceramic |
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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |