CN102737376A - Improved region growing method applied to coronary artery angiography image segmentation - Google Patents
Improved region growing method applied to coronary artery angiography image segmentation Download PDFInfo
- Publication number
- CN102737376A CN102737376A CN201210095033XA CN201210095033A CN102737376A CN 102737376 A CN102737376 A CN 102737376A CN 201210095033X A CN201210095033X A CN 201210095033XA CN 201210095033 A CN201210095033 A CN 201210095033A CN 102737376 A CN102737376 A CN 102737376A
- Authority
- CN
- China
- Prior art keywords
- growth
- seed points
- point
- pixel
- seed
- 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
Images
Abstract
The invention relates to an improved region growing method which is applied to vessel segmentation and extraction in a coronary artery angiography image. The improved region growing method comprises the following steps of: preprocessing the image to obtain an original image capable of directly performing region growth; making a regulation and randomly generating a group of seed points; setting a stack data structure, enabling a newly grown pixel point to enter a stack, and taking out the point previously entering the stack to serve as a current point to be subjected to growth when the current point completes the growth; sequentially performing growth on each seed point, wherein a seed point gray value serves as an average value at a growing initial stage, and calculating a new average gray value when a new pixel point is grown every time along with the growth of the seed points; and completing the growth when no pixel point meeting growth standards exists and no seed point exists. The improved region growing method has the advantages that the seed points are automatically generated, no manual intervention is needed, the local average values around each pixel point serve as growth parameters in a growing process, the coronary artery angiography image with uneven brightness can be segmented, and the efficiency and the accuracy of the image segmentation are improved.
Description
One, technical field
The present invention relates to Flame Image Process, particularly a kind of improved region-growing method is used for the blood vessel segmentation of coronarogram and extracts.
Two, background technology
Medical Image Processing is a branch important in the image processing field, and the segmented extraction of blood vessel, is again one of emphasis and difficult point of Medical Image Processing.The coronarography technology is meant behind the prickle arteriae perforantes inserts conduit, and injects contrast preparation, utilizes X ray different to contrast preparation and other tissue degree of absorption, and the blood vessel image of noting.So brightness of image is very inhomogeneous.As shown in Figure 1, be a coronarogram through smothing filtering, the brightness irregularities of image.
The region growing algorithm is proposed by Zucker the earliest, and as a serial zone algorithm, it comes constantly growth enlarged area since one or one group of initial seed points according to the similarity of pixel in the target area, finally generates the target area.Just it have simply, flexibly, characteristics such as computing velocity is fast, the region growing algorithm application is very extensively.
But the region growing algorithm also exists some problems, like needs artificial selection seed points, particularly when cutting apart the image of uneven brightness, is difficult to the design growth criterion to reach optimum efficiency.When adopting the traditional region growth algorithm to cut apart the coronary angiography image, there are following 2 deficiencies:
1, choosing of seed points needs man-machine interactively, inefficiency.In general, angiogram medium vessels part is maximum owing to contrast preparation absorbs X ray, so the vasculature part gray scale that shows is maximum, general employing is artificial chooses point the brightest among the radiography figure as seed;
2, the growth criterion of traditional area growth is according to neighbor or judge that with the similarity of mean pixel gray-scale value value this criterion is difficult to cut apart the uneven brightness image.Like Fig. 1, the background at A place, also bigger than the gray scale of B place blood vessel, the average gray of the utilization similarity or the overall situation is easy to produce outgrowth or owes growth.Some statistical informations of utilizing pixel are also arranged in addition as the growth criterion, but the calculated amount of those methods increases greatly.
Three, summary of the invention
The present invention seeks to the coronarogram that uneven brightness is arranged to be looked like to cut apart for a kind of image segmentation algorithm more fast and accurately is provided.For realizing above-mentioned purpose, 2 improvement have been proposed on the basis of traditional region growth algorithm.1, according to preset criterion, one group of seed points of automated randomized generation; 2, in the seed points growth course, the pixel grey scale information of utilization regional area so just can be partitioned into the image of uneven brightness as the parameter of growth criterion.The step of enforcement of the present invention comprises:
Image is carried out pre-service, obtain can the deployment area growth algorithm original image;
According to preset criterion, one group of seed points of automated randomized generation;
A stack data structure is set, as the container of growth time point;
Carry out the growth of each seed points successively, in the process of growth, the average gray value that the growth criterion is used is a dynamic calculation in this seed points growth district;
When each seed points all can't regrowth, promptly get result to the end.
2 improvement that the present invention sends out region growing have not only improved the speed of image segmentation, can also improve the accuracy of uneven brightness image segmentation.Particularly suitable is cut apart the coronarogram picture.
Four, description of drawings
Fig. 1 is that the present invention carries out the flow chart of steps that coronarogram looks like to cut apart.
Fig. 2 is the original image that the present invention handles instance
Fig. 3 is that the traditional area growth algorithm is manually chosen the effect that a seed points is handled
Fig. 4 is that the traditional area growth algorithm is manually chosen the effect that a plurality of seed points are handled
Fig. 5 is that the traditional area growth algorithm is chosen the effect that a plurality of seed points are handled automatically
Fig. 6 is the effect that the present invention handles
Five embodiments
The basic thought of region growing is that the collection of pixels with similar quality is got up to constitute the zone.The starting point of a sub pixel as growth looked for to each zone that need cut apart by concrete elder generation, then with having the pixel of same or similar character to merge in the zone at sub pixel place with sub pixel in the field around the sub pixel.These new pixels are used as new sub pixel proceed top process, can include up to the pixel that satisfies condition never again.Know clearly below in conjunction with Fig. 1 and to describe the concrete steps that the present invention carries out image segmentation.
Fig. 2 is a pretreated coronary angiography figure of process, establishes its wide M of being at present, and height is N.(i, j) denotation coordination is that (i, gray values of pixel points j) is to labelled amount F of each point with G
I, j, F
I, j(i j) does not belong to growth district, F to=0 expression point
I, j=1 this point of expression belongs to is grown.
Step 1: the automatic generation of seed points
The traditional region growth method at first need be confirmed one or one group of seed points, wherein mainly leans on man-machine interaction manually to choose.The random seed point is meant and produces some points at random, according to some prioris, corresponding judgment criteria is set; Like gray values of pixel points etc., confirm whether this point satisfies the seed requirement, if satisfy then it is made as seed points; So circulation produces a quantity of seeds point.
Seed points is unsuitable too many, because be to produce at random, seed points can increase calculated amount too much greatly; Seed points can not can be dispersed in the various piece of area-of-interest because will guarantee seed points very little, and through experimental verification, according to feature of image, the seed points number is the 10% ~ 15% proper of area-of-interest pixel sum.Produce one group of seed points S at present at random
1, S
2, S
3S
KK altogether, the gray-scale value of each seed points is G
i, i=1,2 ... K.
Step 2: create stack architexture
The container of the point when creating a stack architexture as growth, the pixel that each growth obtains carries out stack-incoming operation, goes out stack operation after the point of having grown, and obtains the pixel that the next one will be grown.Here use a list structure as stack, define as follows:
typedef?struct?PointStack
{
int?x;
int?y;
struct?PointStack?*next;
}PointStack;
Begin stackedly during growth from first seed points, judge whether the pixel gray-scale value in its neighbours territory meets the growth criterion, the pixel labelled amount that will meet puts 1, and is stacked then.
Step 3: carry out cycling deposition
The local growth criterion is meant in the process mistake of growing, and do not use fixedly mean value and global criteria value as standard value, but according to the moving average of the point that adds in each seed points growth course, as the parameter in the growth criterion.
At first that first seed points is stacked, the regional area of first seed points begins growth.The grey scale pixel value of first seed points as the draw value, is popped, with the pixel that obtains as current point; Judge that whether its neighbours territory pixel satisfies the growth criterion, and whether labelled amount F be 0, if having; Then the F with correspondence is changed to 1, and this point is stacked, and calculates new growth district mean value.Current point satisfies the neighborhood point of growth, and perhaps the neighborhood point disposes, and then goes out stack operation, and the pixel that obtains is as current point, continued growth.
The growth criterion is: with x
Individual seed points be grown to example, initial average gray is the gray-scale value G of seed points
x, when initiate pixel, recomputate average gray value, establishing the zone that x seed points grow into is Ф x, N
xCurrent number of pixels, the then average gray of all pixels in the current region among the expression Ф x
The time represent that this point can join growth district.
Accomplish the growth of first seed points, then that second point is stacked, continued growth, all growing until all seed points finishes, and whole growth process finishes, and all labelled amounts are that 1 pixel promptly is the target area that will cut apart.
Claims (5)
1. one kind is applied in the improved region growing method of coronarogram in looking like to cut apart, and at first the coronary angiography figure that obtains is carried out pre-service, carries out dividing processing then, and the method for dividing processing may further comprise the steps:
(1) lays down a regulation, generate one group of seed points at random;
(2) stack data structure is set;
(3) successively each seed points is grown, initial stage of growth is a mean value with the seed points gray-scale value, and the pixel that every growth makes new advances just calculates new average gray value;
(4) when the pixel that satisfies the growth criterion, and growth ending when not having seed points.
2. according to the method in the claim 1; When selecting one group of seed points at random, need formulate the choosing rule of deleting of seed points in advance, such as according to gray values of pixel points as judgment criteria; When at random to point when satisfying the gray-scale value that is provided with in advance and requiring, then receive this point as seed points.
3. seed points can increase calculated amount too much, but will guarantee to be dispersed in the various piece of area-of-interest, and seed points again can not be very little.
4. through experimental verification, according to feature of image, the seed points number is the 10% ~ 15% proper of area-of-interest pixel sum.
5. according to the method in the claim 1, stack data structure, effect is in growth course; When the pixel that meets the criterion of growing is arranged in the current point neighbours territory, just should point stacked, after the growth of current point is accomplished; Pop, the point of taking-up is as the current point continued growth again;
According to the method in the claim 1, the growth criterion of current point is: with x
Individual seed points be grown to example, initial average gray is the gray-scale value G of seed points
x, when initiate pixel, recomputate average gray value, establishing the zone that x seed points grow into is Ф x, N
xCurrent number of pixels, the then average gray of all pixels in the current region among the expression Ф x
The time represent that this point can join growth district.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201210095033XA CN102737376A (en) | 2012-03-31 | 2012-03-31 | Improved region growing method applied to coronary artery angiography image segmentation |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201210095033XA CN102737376A (en) | 2012-03-31 | 2012-03-31 | Improved region growing method applied to coronary artery angiography image segmentation |
Publications (1)
Publication Number | Publication Date |
---|---|
CN102737376A true CN102737376A (en) | 2012-10-17 |
Family
ID=46992801
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201210095033XA Pending CN102737376A (en) | 2012-03-31 | 2012-03-31 | Improved region growing method applied to coronary artery angiography image segmentation |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN102737376A (en) |
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103810363A (en) * | 2012-11-09 | 2014-05-21 | 上海联影医疗科技有限公司 | Blood vessel seed point selecting method and blood vessel extracting method in angiography |
CN103946868A (en) * | 2013-05-13 | 2014-07-23 | 黄勃 | Processing method and system for medical images |
CN104978725A (en) * | 2014-04-03 | 2015-10-14 | 上海联影医疗科技有限公司 | Method and device for dividing coronary artery |
CN104166979B (en) * | 2013-12-31 | 2017-11-28 | 上海联影医疗科技有限公司 | A kind of vessel extraction method |
CN107705303A (en) * | 2017-10-16 | 2018-02-16 | 长沙乐成医疗科技有限公司 | The dividing method of blood vessel on a kind of medical image |
CN108198239A (en) * | 2017-12-27 | 2018-06-22 | 中山大学 | A kind of three-dimensional visualization method for realizing blood vessel dynamic simulation |
CN109117837A (en) * | 2018-07-26 | 2019-01-01 | 上海鹰瞳医疗科技有限公司 | Area-of-interest determines method and apparatus |
CN109712163A (en) * | 2018-12-05 | 2019-05-03 | 上海联影医疗科技有限公司 | Coronary artery extracting method, device, image processing workstations and readable storage medium storing program for executing |
CN112308846A (en) * | 2020-11-04 | 2021-02-02 | 赛诺威盛科技(北京)有限公司 | Blood vessel segmentation method and device and electronic equipment |
CN112581398A (en) * | 2020-12-22 | 2021-03-30 | 上海电机学院 | Image noise reduction method based on region growing labels |
US10970836B2 (en) | 2013-12-17 | 2021-04-06 | Koninklijke Philips N.V. | Spectral image data processing |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101593275A (en) * | 2009-03-30 | 2009-12-02 | 重庆工学院 | The vena characteristic extracting method of finger vena identification system |
-
2012
- 2012-03-31 CN CN201210095033XA patent/CN102737376A/en active Pending
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101593275A (en) * | 2009-03-30 | 2009-12-02 | 重庆工学院 | The vena characteristic extracting method of finger vena identification system |
Non-Patent Citations (4)
Title |
---|
LI KE ET AL: "An Improved Target Extraction Algorithm Based on Region Growing for Lung CT Image", 《2007 IEEE/ICME INTERNATIONAL CONFERENCE ON COMPLEX MEDICAL ENGINEERING》 * |
史静: "肺部毛细血管分割算法", 《计算机***应用》 * |
夏晶 等: "基于区域生长的前视红外图像分割方法", 《激光与红外》 * |
彭丰平 等: "基于自适应区域生长算法的肝脏分割", 《计算机工程与应用》 * |
Cited By (16)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103810363B (en) * | 2012-11-09 | 2015-07-01 | 上海联影医疗科技有限公司 | Blood vessel seed point selecting method and blood vessel extracting method in angiography |
CN103810363A (en) * | 2012-11-09 | 2014-05-21 | 上海联影医疗科技有限公司 | Blood vessel seed point selecting method and blood vessel extracting method in angiography |
CN103946868A (en) * | 2013-05-13 | 2014-07-23 | 黄勃 | Processing method and system for medical images |
WO2014183246A1 (en) * | 2013-05-13 | 2014-11-20 | Huang Bo | Medical image processing method and system |
US10970836B2 (en) | 2013-12-17 | 2021-04-06 | Koninklijke Philips N.V. | Spectral image data processing |
CN104166979B (en) * | 2013-12-31 | 2017-11-28 | 上海联影医疗科技有限公司 | A kind of vessel extraction method |
CN104978725A (en) * | 2014-04-03 | 2015-10-14 | 上海联影医疗科技有限公司 | Method and device for dividing coronary artery |
CN107705303A (en) * | 2017-10-16 | 2018-02-16 | 长沙乐成医疗科技有限公司 | The dividing method of blood vessel on a kind of medical image |
CN108198239A (en) * | 2017-12-27 | 2018-06-22 | 中山大学 | A kind of three-dimensional visualization method for realizing blood vessel dynamic simulation |
CN108198239B (en) * | 2017-12-27 | 2021-07-23 | 中山大学 | Three-dimensional visualization method for realizing dynamic simulation of blood vessel |
CN109117837A (en) * | 2018-07-26 | 2019-01-01 | 上海鹰瞳医疗科技有限公司 | Area-of-interest determines method and apparatus |
CN109117837B (en) * | 2018-07-26 | 2021-12-07 | 上海鹰瞳医疗科技有限公司 | Region-of-interest determination method and apparatus |
CN109712163A (en) * | 2018-12-05 | 2019-05-03 | 上海联影医疗科技有限公司 | Coronary artery extracting method, device, image processing workstations and readable storage medium storing program for executing |
CN112308846A (en) * | 2020-11-04 | 2021-02-02 | 赛诺威盛科技(北京)有限公司 | Blood vessel segmentation method and device and electronic equipment |
CN112308846B (en) * | 2020-11-04 | 2021-07-13 | 赛诺威盛科技(北京)股份有限公司 | Blood vessel segmentation method and device and electronic equipment |
CN112581398A (en) * | 2020-12-22 | 2021-03-30 | 上海电机学院 | Image noise reduction method based on region growing labels |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN102737376A (en) | Improved region growing method applied to coronary artery angiography image segmentation | |
Anjna et al. | Review of image segmentation technique | |
CN104134221B (en) | Image segmentation method based on self-adaptive genetic algorithm and OTSU algorithm | |
CN109559316A (en) | A kind of improved graph theory dividing method based on super-pixel | |
CN106447020B (en) | A kind of intelligence method for counting colonies | |
CN108108736A (en) | A kind of solar energy photovoltaic panel spot identification method | |
CN109166116A (en) | A kind of image processing algorithm for chip counter | |
CN108596038A (en) | Erythrocyte Recognition method in the excrement with neural network is cut in a kind of combining form credit | |
CN105405136A (en) | Self-adaptive spinal CT image segmentation method based on particle swarm optimization | |
CN104809694A (en) | Digital image processing method and digital image processing device | |
CN107169983A (en) | Multi-threshold image segmentation method based on cross and variation artificial fish-swarm algorithm | |
CN107527364A (en) | A kind of seaweed growing area monitoring method based on remote sensing images and lace curtaining information | |
CN104751200B (en) | A kind of method of SVM network traffic classification | |
CN107274425B (en) | A kind of color image segmentation method and device based on Pulse Coupled Neural Network | |
CN103729621A (en) | Plant leaf image automatic recognition method based on leaf skeleton model | |
Fawzia Rahim et al. | Highly accurate tomato maturity recognition: combining deep instance segmentation, data synthesis and color analysis | |
CN106127763A (en) | One has extensive adaptive image binaryzation method | |
CN103530866B (en) | A kind of image processing method based on Gaussian cloud transformation and device | |
CN102496146B (en) | Image segmentation method based on visual symbiosis | |
CN103955946A (en) | Ant colony neighbor cell tracking system based on cooperation and competition mode and application thereof | |
CN106067175B (en) | A kind of solar activity object detection method based on square net structure | |
CN107705303A (en) | The dividing method of blood vessel on a kind of medical image | |
CN102194224A (en) | Method for identifying tank group in optical remote sensing image | |
CN105405124B (en) | Vessel branch extracting method based on dynamic processing window and two-value area tracking | |
Singh et al. | Automatic seed placement in region growing image segmentation |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20121017 |