CN105205823A - Image segmentation method based on morphology - Google Patents
Image segmentation method based on morphology Download PDFInfo
- Publication number
- CN105205823A CN105205823A CN201510609678.4A CN201510609678A CN105205823A CN 105205823 A CN105205823 A CN 105205823A CN 201510609678 A CN201510609678 A CN 201510609678A CN 105205823 A CN105205823 A CN 105205823A
- Authority
- CN
- China
- Prior art keywords
- image
- carried out
- license plate
- corrosion
- denoising
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- 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
Abstract
The invention discloses an image segmentation method based on morphology and belongs to the technical field of an image segmentation method. The method solves the problem that an image segmentation method in the prior art can not inhibit noise and problems about feature extraction, edge detection, shape recognition, texture analysis, image restoration and reconstruction and other aspects. The method includes the steps that firstly, an original license plate image is acquired; secondly, the original license plate image is preprocessed, that is, filtering and de-noising processing is carried out on the original license plate image; thirdly, the de-noised original license plate image is subjected to morphological algorithm processing; fourthly, each step of image processed through morphology is displayed; fifthly, an image displayed at a threshold after morphological processing and an image displayed at a threshold without morphological processing are compared and analyzed. The method is used for image processing.
Description
Technical field
A kind of based on morphologic image partition method, for image procossing, belong to image partition method technical field.
Background technology
In recent years, China's automobile quantity rapidly increases.Statistics from China Association for Automobile Manufacturers shows, 2008, and China's sale of automobile total amount is 9,380,000, is 1.59 times of sale of automobile total amount in 2005 5900000.Estimate according to China Association for Automobile Manufacturers, within 2009, China's automobile market will increase by maintenance.
Along with the cumulative year after year of automobile quantity, put is huge urban traffic pressure in face of us.How to carry out traffic administration efficiently, more and more become our real-life focal issue.For this problem, people use advanced science and technology, and in succession developed various traffic route supervision, management system, these systems generally all comprise vehicle detection apparatus.And an important technology of the advanced just traffic detection system of the location of car plate and image Segmentation Technology, because it is the early-stage preparations of Car license recognition, and Car license recognition is the technology of core the most in traffic control system.By the identification of car plate to passing vehicle examinations, relevant car plate data can be extracted, for the object reaching monitoring, manage and direct traffic.
License Plate is exactly from the image comprising car plate, adopts image processing techniques to orient the exact position of license plate area.After collecting license plate image, the accuracy rate of License Plate be improved, the accuracy of algorithm of locating license plate of vehicle can only be relied on.In order to improve the accuracy rate of algorithm of locating license plate of vehicle as far as possible, we should combine consideration image acquisition step and License Plate step.Such as we should improve the sharpness collecting image as far as possible, reduce illumination variation to the impact gathering image, make the background of the image collected as far as possible simple, do not comprise the region similar with license plate.If collection in worksite to image in background fairly simple, the area ratio that license plate area accounts for entire image is higher, and in image, car plate geometric distortion does not occur, and License Plate can adopt a step localization method.Namely direct car plate in image to be searched for, orient the position of car plate.If the background of image is complicated, car plate has again certain distortion, then adopt a step localization method to be difficult to obtain the exact boundary of car plate.In this case, we generally will adopt twice localization method, and first algorithm for design Primary Location goes out the position of car plate, then to utilizing Mathematical Morphology Method to carry out the process such as binaryzation, geometry correction to the car plate of just locating.If just location obtains a more than license plate area, must judge these regions, remove pseudo-car plate.Then secondary location is carried out to the car plate of just locating, accurately determine up-and-down boundary and the right boundary of car plate, obtain the accurate positioning result of car plate.
The research of Iamge Segmentation is subject to the great attention of people for many years always, has proposed thousands of kinds of all types of dividing methods so far.Along with the development of each subject, many new Theories and methods are used in Iamge Segmentation by people, obtain the image Segmentation Technology that some are new, comprise the dividing method based on mathematical morphology, the dividing method based on neural network, the dividing method based on wavelet transformation, the dividing method based on somatotype theory, in addition, due to the development of imaging device and technology, people are also deep have studied some special image Segmentation Technology, as dividing methods such as 3-D view, coloured image, texture image, video images.And conventional images segmentation is difficult to the exact boundary obtaining car plate.
Summary of the invention
The present invention is directed to the deficiencies in the prior art part provides a kind of based on morphologic image partition method, solve image partition method of the prior art can not restraint speckle, feature extraction, rim detection, shape recognition, texture analysis, the aspect such as Postprocessing technique and reconstruction problem.
To achieve these goals, the technical solution used in the present invention is:
A kind of based on morphologic image partition method, it is characterized in that, following steps:
(1) former license plate image is obtained;
(2) pre-service is carried out to former license plate image, namely filtering and noise reduction process is carried out to former car plate beginning image;
(3) morphology operations process is carried out to the former license plate image after denoising;
(4) the often step image after Morphological scale-space is shown;
(5) image in threshold value shown after Morphological scale-space and the image in threshold value after not making Morphological scale-space are analyzed.
Further, in described step (3), the step of the original image after denoising being carried out to morphology operations process is:
(31) dilation and corrosion is carried out to the image after denoising;
(32) open and close operator is carried out to the image carried out after dilation and corrosion;
(33) image after carrying out open and close operator to be hit or miss conversion;
(34) to hit or miss conversion after image carry out refinement and skeleton extract.
Further, in described step (31), the concrete steps of the image after denoising being carried out to dilation and corrosion are:
(311) structural element B is used, each pixel of scan image A;
(312) AND-operation is done with the bianry image of structural element and its covering;
(313) if be all 0, this pixel of result images is 0, otherwise is 1;
(314) structural element B is used, each pixel of scan image A;
(315) AND-operation is done with the bianry image of structural element and its covering;
(316) if be all 1, this pixel of result images is 1, otherwise is 0, and the result of corrosion treatment makes original bianry image reduce a circle.
Further, in described step (33), to opening, image after closed operation carries out hitting or miss conversion is expressed as: set A is by 3 subset X, the set of Y and Z composition, the object hit is in A, to find the position of X, if X is enclosed in a wicket W, the local background of the X relevant with W is defined as the difference (W-X) of set, then X can obtain Accurate Curve-fitting location sets in A is to the common factor corroded the supplementary set Ac of A by (W-X) after the corrosion of A by X, this common factor is exactly the position that we will look for, we represent the set be made up of the background of X and X by set B, we can make B=(B1, B2), here B1=X, B2=(W-X), then in A, carry out coupling to B can be expressed as: A ⊙ B.
Compared with prior art, the invention has the advantages that:
One, splitting license plate image based on the method for mathematical morphology is use certain structural element, the opening operation in mathematical morphology and closed operation is utilized to process image, obtaining multiple may be the region of car plate, then in image after treatment with multizone diagnostic method multiple may be the correct position finding car plate in the region of car plate;
Two, mathematical morphology has unique advantage in Image Edge-Detection, it is based on set operation, there is nonlinear characteristic, rim detection can either embody image collection feature, good detected image edge, again can requirement of real time, also solve the coordination problem of rim detection precision and noise robustness.
Accompanying drawing explanation
Fig. 1 is framework schematic flow sheet of the present invention.
Embodiment
Below in conjunction with drawings and Examples, the present invention is further illustrated.
A kind of based on morphologic image partition method, it is characterized in that, following steps:
(1) former license plate image is obtained;
(2) pre-service is carried out to former license plate image, namely filtering and noise reduction process is carried out to former car plate beginning image;
(3) morphology operations process is carried out to the former license plate image after denoising; The step of the original image after denoising being carried out to morphology operations process is:
(31) dilation and corrosion is carried out to the image after denoising; The concrete steps of the image after denoising being carried out to dilation and corrosion are:
(311) structural element B is used, each pixel of scan image A;
(312) AND-operation is done with the bianry image of structural element and its covering;
(313) if be all 0, this pixel of result images is 0, otherwise is 1;
(314) structural element B is used, each pixel of scan image A;
(315) AND-operation is done with the bianry image of structural element and its covering;
(316) if be all 1, this pixel of result images is 1, otherwise is 0, and the result of corrosion treatment makes original bianry image reduce a circle.
clearall，closeall
I＝imread(′texttif′)；
SE＝[0，1，0；1，1，1；0，1，0]
BW1＝imdilate(I，SE)；
BW2＝imerode(I，SE)；
figure(1)，
subplot(1，2，1)，imshow(I，′notruesize′)，title(′OriginalImage′)；
subplot(1，2，2)，imshow(BW1，′notruesize′)，title(′DilatedImage′)；
figure(2)，
subplot(1，2，1)，imshow(I，′notruesize′)，title(′OriginalImage′)；
subplot(1，2，2)，imshow(BW2，′notruesize′)，title(′ErodedImage′)；
(32) open and close operator is carried out to the image carried out after dilation and corrosion;
clearall，closeall
I＝imread(′nodules1.GIF′)；
bw＝～im2bw(I，graythresh(I))；
se＝strel(′disk′，5)；
bw2＝imopen(bw，se)；
subplot(1，2，1)，imshow(bw)，title(′ThresholdedImage′)
subplot(1，2，2)，imshow(bw2)，title(′Afteropening′)
bw3＝imclose(bw，se)；
figure；
subplot(1，2，1)，imshow(bw，′truesize′)，title(′ThresholdedImage′)
subplot(1，2，2)，imshow(bw3，′truesize′)，title(′AfterClosing′)
(33) image after carrying out open and close operator to be hit or miss conversion, to opening, image after closed operation carries out hitting or miss conversion is expressed as: set A is by 3 subset X, the set of Y and Z composition, the object hit is in A, to find the position of X, if X is enclosed in a wicket W, the local background of the X relevant with W is defined as the difference (W-X) of set, then X can obtain Accurate Curve-fitting location sets in A is to the common factor corroded the supplementary set Ac of A by (W-X) after the corrosion of A by X, this common factor is exactly the position that we will look for, we represent the set be made up of the background of X and X by set B, we can make B=(B1, B2), here B1=X, B2=(W-X), then in A, carry out coupling to B can be expressed as: A
⊙B。
bw2＝bwhitmiss(bw，interval)
subplot(1，3，1)，imshow(bw，′notruesize′)，title(′OriginalImage′)；
subplot(1，3，2)，imshow(interval，′truesize′)，title(′IntervalImage′)；
subplot(1，3，3)，imshow(bw2，′notruesize′)，title(′afterhit/misstransformation′)；
(34) to hit or miss conversion after image carry out refinement and skeleton extract.
clearall，closeall
BW＝～imread(′logo.GIF′)；
BW1＝bwmorph(BW，′thin′，Inf)；
BW2＝bwmorph(BW，′skel′，Inf)；
subplot(1，3，1)，imshow(BW)，title(′OriginalImage′)；
subplot(1，3，2)，imshow(BW1)，title(′ThinnedImage′)；
subplot(1，3，3)，imshow(BW2)，title(′Imageskeleton′)；
(4) the often step image after Morphological scale-space is shown;
(5) image in threshold value shown after Morphological scale-space and the image in threshold value after not making Morphological scale-space are analyzed.
Claims (4)
1. based on a morphologic image partition method, it is characterized in that, following steps:
(1) former license plate image is obtained;
(2) pre-service is carried out to former license plate image, namely filtering and noise reduction process is carried out to former car plate beginning image;
(3) morphology operations process is carried out to the former license plate image after denoising;
(4) the often step image after Morphological scale-space is shown;
(5) image in threshold value shown after Morphological scale-space and the image in threshold value after not making Morphological scale-space are analyzed.
2. one according to claim 1 is based on morphologic image partition method, it is characterized in that: in described step (3), and the step of the original image after denoising being carried out to morphology operations process is:
(31) dilation and corrosion is carried out to the image after denoising;
(32) open and close operator is carried out to the image carried out after dilation and corrosion;
(33) image after carrying out open and close operator to be hit or miss conversion;
(34) to hit or miss conversion after image carry out refinement and skeleton extract.
3. one according to claim 1 is based on morphologic image partition method, it is characterized in that: in described step (31), and the concrete steps of the image after denoising being carried out to dilation and corrosion are:
(311) structural element B is used, each pixel of scan image A;
(312) AND-operation is done with the bianry image of structural element and its covering;
(313) if be all 0, this pixel of result images is 0, otherwise is 1;
(314) structural element B is used, each pixel of scan image A;
(315) AND-operation is done with the bianry image of structural element and its covering;
(316) if be all 1, this pixel of result images is 1, otherwise is 0, and the result of corrosion treatment makes original bianry image reduce a circle.
4. one according to claim 1 is based on morphologic image partition method, it is characterized in that: in described step (33), to opening, image after closed operation carries out hitting or miss conversion is expressed as: set A is by 3 subset X, the set of Y and Z composition, the object hit is in A, to find the position of X, if X is enclosed in a wicket W, the local background of the X relevant with W is defined as the difference (W-X) of set, then X can obtain Accurate Curve-fitting location sets in A is to the common factor corroded the supplementary set Ac of A by (W-X) after the corrosion of A by X, this common factor is exactly the position that we will look for, we represent the set be made up of the background of X and X by set B, we can make B=(B1, B2), here B1=X, B2=(W-X), then in A, carry out coupling to B can be expressed as: A ⊙ B.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510609678.4A CN105205823A (en) | 2015-09-23 | 2015-09-23 | Image segmentation method based on morphology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510609678.4A CN105205823A (en) | 2015-09-23 | 2015-09-23 | Image segmentation method based on morphology |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105205823A true CN105205823A (en) | 2015-12-30 |
Family
ID=54953484
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510609678.4A Pending CN105205823A (en) | 2015-09-23 | 2015-09-23 | Image segmentation method based on morphology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105205823A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108876784A (en) * | 2018-06-27 | 2018-11-23 | 清华大学 | A kind of image processing method and device removing flat work pieces connecting component |
| CN109816847A (en) * | 2018-12-20 | 2019-05-28 | 深圳怡化电脑股份有限公司 | A kind of method, apparatus and terminal device judging that hand-written writing is altered |
| CN114549811A (en) * | 2021-12-31 | 2022-05-27 | 哈尔滨理工大学 | License plate identification method based on license plate spacer body characteristics |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101246551A (en) * | 2008-03-07 | 2008-08-20 | 北京航空航天大学 | Fast license plate locating method |
| CN101789080A (en) * | 2010-01-21 | 2010-07-28 | 上海交通大学 | Detection method for vehicle license plate real-time positioning character segmentation |
-
2015
- 2015-09-23 CN CN201510609678.4A patent/CN105205823A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101246551A (en) * | 2008-03-07 | 2008-08-20 | 北京航空航天大学 | Fast license plate locating method |
| CN101789080A (en) * | 2010-01-21 | 2010-07-28 | 上海交通大学 | Detection method for vehicle license plate real-time positioning character segmentation |
Non-Patent Citations (6)
| Title |
|---|
| 佚名: "OpenCV形态学操作", 《HTTP://BLOG.CSDN.NET/BYXDAZ/ARTICLE/DETAILS/5775717》 * |
| 佚名: "数字图像处理课程设计", 《HTTPS://WENKU.BAIDU.COM/VIEW/6BAD2A01B84AE45C3A358C2C.HTML》 * |
| 卢雅琴等: "基于数学形态学的车牌定位方法", 《计算机工程》 * |
| 廖春生: "基于形态学车牌定位算法仿真研究", 《计算机仿真》 * |
| 张涛等: "《MATLAB图像处理编程与应用》", 30 April 2014 * |
| 王昭莲等: "基于数学形态学的机场跑道快速识别方法", 《计算机技术与发展》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108876784A (en) * | 2018-06-27 | 2018-11-23 | 清华大学 | A kind of image processing method and device removing flat work pieces connecting component |
| CN108876784B (en) * | 2018-06-27 | 2021-07-30 | 清华大学 | Image processing method and device for removing connecting part of planar workpiece |
| CN109816847A (en) * | 2018-12-20 | 2019-05-28 | 深圳怡化电脑股份有限公司 | A kind of method, apparatus and terminal device judging that hand-written writing is altered |
| CN109816847B (en) * | 2018-12-20 | 2021-01-01 | 深圳怡化电脑股份有限公司 | Method and device for judging handwritten handwriting correction and terminal equipment |
| CN114549811A (en) * | 2021-12-31 | 2022-05-27 | 哈尔滨理工大学 | License plate identification method based on license plate spacer body characteristics |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Aggarwal et al. | Feature extraction and LDA based classification of lung nodules in chest CT scan images | |
| CN101551863B (en) | Method for extracting roads from remote sensing image based on non-sub-sampled contourlet transform | |
| CN102842136B (en) | A kind of optic disk projective iteration method of comprehensive vascular distribution and optic disk appearance characteristics | |
| CN103679714B (en) | A kind of optics and SAR automatic image registration method based on gradient cross-correlation | |
| CN105740945A (en) | People counting method based on video analysis | |
| CN102043954B (en) | Quick and robust iris recognition method based on related function matching | |
| CN102542556B (en) | Method for automatically extracting ultrasonic breast tumor image | |
| CN104299009B (en) | License plate character recognition method based on multi-feature fusion | |
| CN104933711A (en) | Automatic fast segmenting method of tumor pathological image | |
| CN106709950A (en) | Binocular-vision-based cross-obstacle lead positioning method of line patrol robot | |
| CN103345617A (en) | Method and system for recognizing traditional Chinese medicine | |
| CN106127205A (en) | A kind of recognition methods of the digital instrument image being applicable to indoor track machine people | |
| CN106815583A (en) | A kind of vehicle at night license plate locating method being combined based on MSER and SWT | |
| CN105205823A (en) | Image segmentation method based on morphology | |
| CN107480585A (en) | Object detection method based on DPM algorithms | |
| CN109687382B (en) | Relay protection pressing plate switching state identification method based on color template matching | |
| CN116823811B (en) | Functional jacket surface quality detection method | |
| CN102073872A (en) | Image-based method for identifying shape of parasite egg | |
| CN104298993B (en) | A kind of bar number positioning and recognition methods suitable under complex scene along track | |
| CN104933723A (en) | Tongue image segmentation method based on sparse representation | |
| CN102201060B (en) | Method for tracking and evaluating nonparametric outline based on shape semanteme | |
| CN113436102A (en) | Method for detecting surface defects of machined part | |
| CN107610094A (en) | Based on the three-dimensional crack 3 D detection method characterized of ellipsoid | |
| CN106023166B (en) | The detection method and device of dangerous object hidden by human body in microwave image | |
| CN106355576A (en) | SAR image registration method based on MRF image segmentation algorithm |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20151230 |