CN105300998A - Paper defect detection method based on bit planes - Google Patents
Paper defect detection method based on bit planes Download PDFInfo
- Publication number
- CN105300998A CN105300998A CN201510641399.6A CN201510641399A CN105300998A CN 105300998 A CN105300998 A CN 105300998A CN 201510641399 A CN201510641399 A CN 201510641399A CN 105300998 A CN105300998 A CN 105300998A
- Authority
- CN
- China
- Prior art keywords
- bit plane
- image
- paper
- bit
- gray level
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Landscapes
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a paper defect detection method based on bit planes. The paper defect detection method comprises the following steps of firstly, obtaining an original image of the tested paper, and converting into a grayscale image; pretreating the grayscale image, so as to eliminate the noise in the image; decomposing the bit plane of the pretreated grayscale image to obtain eight bit planes of the image; using Gray code to enhance the bit planes, so as to obtain eight enhanced bit planes of the grayscale image; finally, selecting the sixth enhanced bit plane, and segmenting the image, so as to obtain the final detection result. The paper defect detection method has the advantages that while the rapidity of algorithm is guaranteed, the defect can be well detected; the anti-interference property and positioning accuracy are better, and the operation is simple.
Description
Technical field
The present invention relates to a kind of paper defect testing method, be specifically related to a kind of paper defect testing method based on bit plane.
Background technology
Along with improving constantly of paper machine speed, paper is faced with the risk of the more defects of appearance.Paper defect due to artificial cognition paper needs to drop into a large amount of manpowers, and discrimination is low, inefficiency, so it is impossible to rely on the paper defect of artificial naked eyes to paper to detect.Detect by the paper defect of machine vision to paper and become irreplaceable trend.
The method utilizing the paper defect of machine vision to paper to detect at present is generally divided into threshold method, morphological method, grey level statistics method three class.Wherein threshold method arranges different threshold values according to different paper defects, it is a kind of simple and effective image partition method, but the method is all different for the threshold value selected by different paper, different paper defects, so poor universality, and noise immunity is also not fully up to expectations; The edge detection operator detection paper defects edge that morphological method adopts corrosion and expands, which overcome the tedious work that in defects detection, different paper defects selected threshold is different, and to image detail and edge local, there is good result, particularly to containing noisy image, if but only use Mathematical Morphology Method, still there will be the situation of marginal information loss; Grey level statistics method utilizes the statistical nature of paper defects image to detect paper defects, method based on grey level statistics is of a great variety, and more representative has: one dimension autoregression algorithm, fuzzy logic algorithm and the paper defect testing method etc. based on co-occurrence matrix and self organizing neural network.One dimension autoregression algorithm can not be used for texture modeling and defects detection, fuzzy logic algorithm is for the identification of defect and subsequent treatment more complicated, and very large on the impact of testing result based on the selection of the paper defect testing method degree of confidence of co-occurrence matrix and self organizing neural network.
Summary of the invention
The object of the present invention is to provide a kind of paper defect testing method based on bit plane, to overcome the defect that above-mentioned prior art exists, the present invention while guarantee algorithm rapidity, can detect defect preferably, there is good anti-interference and Position location accuracy, and computing is simple.
For achieving the above object, the present invention adopts following technical scheme:
Based on a paper defect testing method for bit plane, comprise the following steps:
Step 1: the original image obtaining tested paper, and be converted to gray level image;
Step 2: carry out pre-service to gray level image, with the noise in removal of images;
Step 3: pretreated gray level image is carried out eight bit planes that Bit Plane Decomposition obtains image;
Step 4: adopt Gray code bitplanes to strengthen, eight that obtain gray level image strengthen bit plane;
Step 5: choose the 6th and strengthen bit plane, Iamge Segmentation is carried out to it, obtains final testing result.
Further, in step 2, pretreated method is carried out for gray level image is carried out adaptive median filter process to gray level image.
Further, adaptive median filter process comprises the following steps:
Step 2.1: first determine maximum filter radius r
maxwith initial filter radius r
0;
Step 2.2: start filtering, calculates the neighborhood intermediate value I of current pixel I (x, y)
med, neighborhood maximums I
maxwith neighborhood minimum I
min;
Step 2.3: judge current neighborhood intermediate value I
medwhether meet I
min< I
med< I
maxif meet, then current neighborhood intermediate value is not noise spot, enters step 2.4, otherwise filter radius adds 1, and gets back to step 2.2 and continue filtering, if filter radius equals maximum filter radius, then forwards step 2.4 to;
Step 2.4: judge whether current pixel point I (x, y) meets I
min< I (x, y) < I
maxif meet, then current pixel point is not noise spot, current pixel value initial value exports, otherwise, with current neighborhood intermediate value I
medsubstitute current pixel value I (x, y) to export.
Further, the maximum filter radius in step 2.1 is 10, and initial filter radius is 3.
Further, the method for in step 3, pretreated gray level image being carried out Bit Plane Decomposition is: the identical binary bit extracting each pixel respectively forms a plane, and the gray level image being about to have 256 gray levels is decomposed into eight bit planes.
Further, the method adopting Gray code bitplanes to strengthen in step 4 is: adopt following formula:
In formula,
for xor operation, a
ifor i-th bit plane that Bit Plane Decomposition obtains, g
ibit plane a
igray code represent, namely i-th strengthen bit plane.
Further, in step 5 to the method that the 6th enhancing bit plane carries out Iamge Segmentation be: set F as the 6th enhancing bit plane, B to be size be 3 × 3 square structure element, first F is allowed to be corroded by B, then the difference of image F and its corrosion is asked for, if D is final Edge detected image, utilize formula as follows:
D=F-(FΘB)
In formula, Θ is etching operation, and F is the image at edge to be extracted, and namely the 6th strengthens bit plane, and B is the square structure element of 3 × 3, and D is the final edge image detected.
Compared with prior art, the present invention has following useful technique effect:
The inventive method is simple and easy to implement, and when not affecting production, effectively can detect paper defects, having good anti-interference and Position location accuracy, and computing is simple, thus reaches raising papermaking industry automaticity, reduces manually-operated object.By the paper defect testing method based on bit plane of the present invention, the edge detected is more complete, and profile is more clear, does not have false edge, and the interference by background impurities and noise is less, and be simple and easy to implement, calculated amount is few.If apply the present invention to commercial paper scene, traditional detection method poor anti jamming capability can be solved preferably, locate inaccurate and that computing is complicated problem, thus promote the development of papermaking industry, there is very large market potential.
Accompanying drawing explanation
Fig. 1 is process flow diagram of the present invention.
Embodiment
Below in conjunction with accompanying drawing, the present invention is described in further detail:
The present invention is using the image containing paper defects as detected object, and as shown in Figure 1, concrete implementation step is as follows for treatment scheme:
Step1, obtained the original image of tested paper by CCD camera, and be converted to gray level image.
Step2, adaptive median filter process is carried out to the gray level image obtained, with the noise in removal of images, it can adjust the size of filter window and output filtering result according to image by the annoyance level of noise adaptively, and choosing initial filter radius is 3, and maximum filter radius is 10.Concrete grammar is from the upper left corner of original image, utilize filter window to carry out slip scan, judge whether to there is noise spot, basis for estimation is whether current neighborhood intermediate value in filter window is in current neighborhood tonal range, if not in its neighborhood tonal range, then there is noise, Mesophyticum in filter window is then utilized to replace the value of noise spot, otherwise, then perform initial value and export.Filtering has following key step:
1, first maximum filter radius r to be determined
maxwith initial filter radius r
0;
2, start filtering, calculate the neighborhood intermediate value I of current pixel I (x, y)
med, neighborhood maximums I
maxwith neighborhood minimum I
min;
3, current neighborhood intermediate value I is judged
medwhether meet I
min< I
med< I
maxif meet, then current neighborhood intermediate value is not noise spot, enters 4, otherwise filter radius adds 1, and gets back to 2 continuation filtering, if filter radius has been added to maximum filter radius, then forwards 4 to;
4, judge whether current pixel point I (x, y) meets I
min< I (x, y) < I
maxif meet, then current pixel point is not noise spot, current pixel value initial value exports, otherwise, with current neighborhood intermediate value I
medsubstitute current pixel value I (x, y).
Step3, carry out Bit Plane Decomposition to filtered image, the identical binary bit extracting each pixel respectively forms a plane, namely extracts the b of all pixels
0position forms first plane, extracts the b of all pixels
1position forms second plane, and the rest may be inferred, and this image with 256 gray levels just can be broken down into 8 bit planes.
Step4, employing Gray code are improved the bit plane obtained in Step3, and be enhanced bit plane, reduces the impact of the change bitplanes of gray-scale value with this.Concrete grammar is as shown by the equation:
In formula,
for xor operation, a
ifor i-th bit plane that Bit Plane Decomposition obtains, g
ibit plane a
igray code represent, namely i-th strengthen bit plane.
Compared to bit plane, the image complexity strengthening bit plane reduces, and the bit plane with the information of vision meaning is more.6th strengthens bit plane is the XOR of the 6th and the 7th bit plane of original position plane, includes information the most useful in paper defects image, so choose the 6th enhancing bit plane as experimental subjects in experiment below.
Step5, in conjunction with mathematical morphology, use morphologic edge detection method the 6th of detecting in Step4 to strengthen the edge of bit plane, reach the object of Iamge Segmentation, thus obtain final testing result.
The algorithm utilizing mathematical morphology to extract image border is: be provided with image F, B is a suitable structural element, first allows F be corroded by B, then asks for the difference of image F and its corrosion, if D is the final edge image detected, utilizes formula as follows:
D=F-(FΘB)
In formula, Θ is etching operation, and F is the image at edge to be extracted, and namely the 6th strengthens bit plane, and B is the square structure element of 3 × 3, and D is the final edge image detected.
Claims (7)
1., based on a paper defect testing method for bit plane, it is characterized in that, comprise the following steps:
Step 1: the original image obtaining tested paper, and be converted to gray level image;
Step 2: carry out pre-service to gray level image, with the noise in removal of images;
Step 3: pretreated gray level image is carried out eight bit planes that Bit Plane Decomposition obtains image;
Step 4: adopt Gray code bitplanes to strengthen, eight that obtain gray level image strengthen bit plane;
Step 5: choose the 6th and strengthen bit plane, Iamge Segmentation is carried out to it, obtains final testing result.
2. a kind of paper defect testing method based on bit plane according to claim 1, is characterized in that, carry out pretreated method for gray level image is carried out adaptive median filter process in step 2 to gray level image.
3. a kind of paper defect testing method based on bit plane according to claim 2, it is characterized in that, adaptive median filter process comprises the following steps:
Step 2.1: first determine maximum filter radius r
maxwith initial filter radius r
0;
Step 2.2: start filtering, calculates the neighborhood intermediate value I of current pixel I (x, y)
med, neighborhood maximums I
maxwith neighborhood minimum I
min;
Step 2.3: judge current neighborhood intermediate value I
medwhether meet I
min< I
med< I
maxif meet, then current neighborhood intermediate value is not noise spot, enters step 2.4, otherwise filter radius adds 1, and gets back to step 2.2 and continue filtering, if filter radius equals maximum filter radius, then forwards step 2.4 to;
Step 2.4: judge whether current pixel point I (x, y) meets I
min< I (x, y) < I
maxif meet, then current pixel point is not noise spot, current pixel value initial value exports, otherwise, with current neighborhood intermediate value I
medsubstitute current pixel value I (x, y) to export.
4. a kind of paper defect testing method based on bit plane according to claim 3, it is characterized in that, the maximum filter radius in step 2.1 is 10, and initial filter radius is 3.
5. a kind of paper defect testing method based on bit plane according to claim 1, it is characterized in that, the method of in step 3, pretreated gray level image being carried out Bit Plane Decomposition is: the identical binary bit extracting each pixel respectively forms a plane, and the gray level image being about to have 256 gray levels is decomposed into eight bit planes.
6. a kind of paper defect testing method based on bit plane according to claim 1, it is characterized in that, the method adopting Gray code bitplanes to strengthen in step 4 is: adopt following formula:
In formula,
for xor operation, a
ifor i-th bit plane that Bit Plane Decomposition obtains, g
ibit plane a
igray code represent, namely i-th strengthen bit plane.
7. a kind of paper defect testing method based on bit plane according to claim 1, it is characterized in that, in step 5 to the method that the 6th enhancing bit plane carries out Iamge Segmentation be: set F as the 6th enhancing bit plane, B to be size be 3 × 3 square structure element, first F is allowed to be corroded by B, then ask for the difference of image F and its corrosion, if D is the final edge image detected, utilize formula as follows:
D=F-(FΘB)
In formula, Θ is etching operation, and F is the image at edge to be extracted, and namely the 6th strengthens bit plane, and B is the square structure element of 3 × 3, and D is the final edge image detected.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510641399.6A CN105300998A (en) | 2015-09-30 | 2015-09-30 | Paper defect detection method based on bit planes |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510641399.6A CN105300998A (en) | 2015-09-30 | 2015-09-30 | Paper defect detection method based on bit planes |
Publications (1)
Publication Number | Publication Date |
---|---|
CN105300998A true CN105300998A (en) | 2016-02-03 |
Family
ID=55198506
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201510641399.6A Pending CN105300998A (en) | 2015-09-30 | 2015-09-30 | Paper defect detection method based on bit planes |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN105300998A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107798674A (en) * | 2017-09-15 | 2018-03-13 | 华南理工大学 | A kind of method for removing of surface of workpiece structure flase drop result |
CN110349139A (en) * | 2019-07-02 | 2019-10-18 | 芜湖启迪睿视信息技术有限公司 | A kind of miscellaneous detection method of packaging based on neural network segmentation |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2007122274A (en) * | 2005-10-26 | 2007-05-17 | Sumitomo Chemical Co Ltd | Image analysis device, image analysis method and program for image analysis processing |
CN102411777A (en) * | 2011-06-15 | 2012-04-11 | 夏东 | Method for detecting scratch defects of printing product |
CN104102911A (en) * | 2014-07-09 | 2014-10-15 | 宁波摩视光电科技有限公司 | Image processing for AOI (automated optical inspection)-based bullet appearance defect detection system |
-
2015
- 2015-09-30 CN CN201510641399.6A patent/CN105300998A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2007122274A (en) * | 2005-10-26 | 2007-05-17 | Sumitomo Chemical Co Ltd | Image analysis device, image analysis method and program for image analysis processing |
CN102411777A (en) * | 2011-06-15 | 2012-04-11 | 夏东 | Method for detecting scratch defects of printing product |
CN104102911A (en) * | 2014-07-09 | 2014-10-15 | 宁波摩视光电科技有限公司 | Image processing for AOI (automated optical inspection)-based bullet appearance defect detection system |
Non-Patent Citations (5)
Title |
---|
RAFAEL C. GONZALEZ 等: "《数字图像处理的MATLAB实现》", 30 April 2013, 清华大学出版社,第一版 * |
S.PRIYA 等: ""A Novel Approach to Fabric Defect Detection Using Digital Image Processing", 《PROCEEDINGS OF 2011 INTERNATIONAL CONFERENCE ON SIGNAL PROCESSING, COMMUNICATION, COMPUTING AND NETWORKING TECHNOLOGIES (ICSCCN 2011)》 * |
李俊山,李旭辉: "《数字图像处理》", 30 November 2013, 清华大学出版社,第二版 * |
秦武 等: "基于位平面分解的E∞自动提取去雾算法", 《三峡大学学报(自然科学版)》 * |
韩晓微 等: "《数字图像融合技术》", 31 December 2012, 东北大学出版社,第一版 * |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107798674A (en) * | 2017-09-15 | 2018-03-13 | 华南理工大学 | A kind of method for removing of surface of workpiece structure flase drop result |
CN107798674B (en) * | 2017-09-15 | 2021-04-23 | 华南理工大学 | Method for eliminating false detection result of surface structure of metal workpiece |
CN110349139A (en) * | 2019-07-02 | 2019-10-18 | 芜湖启迪睿视信息技术有限公司 | A kind of miscellaneous detection method of packaging based on neural network segmentation |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN112819772B (en) | High-precision rapid pattern detection and recognition method | |
CN106780486B (en) | Steel plate surface defect image extraction method | |
CN106682665B (en) | Seven-segment type digital display instrument number identification method based on computer vision | |
CN103198319B (en) | For the blurred picture Angular Point Extracting Method under the wellbore environment of mine | |
CN109241973B (en) | Full-automatic soft segmentation method for characters under texture background | |
CN116309600B (en) | Environment-friendly textile quality detection method based on image processing | |
Er-Sen et al. | An adaptive edge-detection method based on the canny operator | |
CN105447489B (en) | A kind of character of picture OCR identifying system and background adhesion noise cancellation method | |
CN108335294A (en) | The power distribution room abnormality image-recognizing method of complex condition | |
CN110687122A (en) | Method and system for detecting surface cracks of ceramic tile | |
CN103325123A (en) | Image edge detection method based on self-adaptive neural fuzzy inference systems | |
CN111369570A (en) | Multi-target detection tracking method for video image | |
CN114943744A (en) | Edge detection method based on local Otsu thresholding | |
CN102509265B (en) | Digital image denoising method based on gray value difference and local energy | |
CN105300998A (en) | Paper defect detection method based on bit planes | |
CN102313740B (en) | Solar panel crack detection method | |
CN106408533A (en) | Card image extraction method and card image extraction system | |
CN116630321B (en) | Intelligent bridge health monitoring system based on artificial intelligence | |
CN104102911A (en) | Image processing for AOI (automated optical inspection)-based bullet appearance defect detection system | |
CN108205678A (en) | A kind of nameplate Text region processing method containing speck interference | |
CN110047041B (en) | Space-frequency domain combined traffic monitoring video rain removing method | |
CN112288726A (en) | Method for detecting foreign matters on belt surface of underground belt conveyor | |
CN102521804A (en) | Image edge detection method based on window modified multi-level median filtering | |
CN106384103A (en) | Vehicle face recognition method and device | |
CN105825514A (en) | Shear wave coefficient probability density distribution-based image edge detection method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20160203 |
|
RJ01 | Rejection of invention patent application after publication |