CN111127479A - Level set image segmentation method based on curve area - Google Patents

Level set image segmentation method based on curve area Download PDF

Info

Publication number
CN111127479A
CN111127479A CN201911301075.2A CN201911301075A CN111127479A CN 111127479 A CN111127479 A CN 111127479A CN 201911301075 A CN201911301075 A CN 201911301075A CN 111127479 A CN111127479 A CN 111127479A
Authority
CN
China
Prior art keywords
curve
area
image
term
energy
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
Application number
CN201911301075.2A
Other languages
Chinese (zh)
Inventor
贺建峰
陈路达
管观华
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Kunming University of Science and Technology
Original Assignee
Kunming University of Science and Technology
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Kunming University of Science and Technology filed Critical Kunming University of Science and Technology
Priority to CN201911301075.2A priority Critical patent/CN111127479A/en
Publication of CN111127479A publication Critical patent/CN111127479A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • 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
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10072—Tomographic images
    • G06T2207/10081—Computed x-ray tomography [CT]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30004—Biomedical image processing
    • G06T2207/30061—Lung

Abstract

The invention relates to a level set image segmentation method based on curve area, and belongs to the technical field of image processing. The method avoids the problems of large calculation amount and low efficiency caused by the fact that only gray information of inner and outer areas of a curve is considered and edge information of the curve is not considered in the calculation motion process of a set curve in a gray uniform image due to a traditional level set segmentation model (CV model). According to the improved method provided by the invention, a curve area item which is fitted based on image edge information is added in the CV model, and the image with uniform gray scale is segmented. Experimental results show that the improved method can obviously improve the calculation and the movement speed of the curve in the CV model and improve the image segmentation efficiency.

Description

Level set image segmentation method based on curve area
Technical Field
The invention relates to a level set image segmentation method based on curve area, and belongs to the technical field of image processing.
Background
Image segmentation is to divide an image into meaningful parts according to a certain uniformity (or consistency) principle, so that each part meets a certain consistency requirement. The traditional level set segmentation CV model utilizes global image information to drive the curve evolution. The driving curve motion is mainly the energy difference between the inside and the outside of the curve, when the boundary of the target is detected, the curve only considers the gray information of the inside and the outside of the curve, but does not consider the edge information of the curve, so that the calculation amount is large, and the efficiency is low. In order to improve the accuracy of the CV model for detecting the target edge, the invention adds a curve area item based on edge information in the traditional CV model, can accelerate the motion speed of the curve and improve the calculation efficiency.
Disclosure of Invention
The invention aims to provide a level set image segmentation method based on curve area, which is used for accelerating the movement speed of a set curve in an image, improving the calculation efficiency and acquiring a segmented image of a region of interest.
The technical scheme of the invention is as follows: a level set image segmentation method based on curve area comprises the following specific steps:
step 1: and drawing a closed curve for any image to be segmented to divide the image into an inner area and an outer area.
Step 2: and calculating the area energy of the whole image region and the area energy inside the closed curve, wherein the area energy outside the closed curve is equal to the area energy of the whole image region minus the area energy inside the closed curve.
Step 3: an image curve area energy constraint term is calculated that is equal to the area energy outside the closed curve minus the area energy inside the closed curve.
Step 4: calculating an overall image energy general function which is equal to the sum of 4 terms; the first term is the length term of the closed curve, the second term is the curve area energy constraint term, and the third and fourth terms are the inner area energy term and the outer area energy term, respectively.
Step 5: and when the closed evolution curve is positioned at the boundary of the target to be segmented, the total energy general function of the image is minimum, and the segmented image is obtained.
The Step2 is specifically as follows:
(1): calculating the area energy S of the whole image regionu，
Figure BDA0002321801810000011
Wherein u is0Representing the gray value of any image to be segmented, wherein H (phi) is a Heaviside function, phi is a level set function, and phi is an edge detection function
Figure BDA0002321801810000012
▽ denotes the gradient operator, GσGaussian function representing window size σ
Figure BDA0002321801810000013
u is a constant and Ω is the image space u0One set of (a).
(2): calculating the area energy S inside the closed curveinside(C)，
Figure BDA0002321801810000021
(3): calculating the area energy S outside the closed curveoutside(C)，
Soutside＝Su-Sinside＝∫Ωg*u0*H(φ)-g*H(φ)dxdy。
The Step3 is specifically as follows:
calculating an image curve area energy constraint term Econv，
Figure BDA0002321801810000022
The Step4 is specifically as follows:
calculating the energy general function of the whole image:
Figure BDA0002321801810000023
Figure BDA0002321801810000024
wherein, the first term of the equation is the length term of the curve C, mu is more than or equal to 0 and is the coefficient of the length term, the second term of the equation is the area term of the curve C with the edge information fused, u is the area term of the curve C0Representing the gray value of any image to be segmented, v is more than or equal to 0 and is the coefficient of area term, lambda1,λ2A coefficient of energy terms in inner and outer regions, c1,c2Respectively representing the mean grey value inside the evolution curve and the mean grey value outside the evolution curve, deltaεAnd (phi) represents the Dirac function.
The invention has the beneficial effects that: the method avoids the situations of large calculation amount and low efficiency caused by the fact that only gray information of inner and outer areas of a curve is considered and edge information of the curve is not considered in the calculation motion process of a set curve in an image due to the traditional level set segmentation CV model. The improved method provided by the invention can obviously improve the calculation and the movement speed of the curve in the CV model and improve the image segmentation efficiency.
Drawings
FIG. 1 is a flow chart of the steps of the present invention;
fig. 2 is a segmented contrast image of a lung CT scan in an embodiment of the present invention, in which (a) is an original image, (b) is an image segmentation of a CV model, and (c) is an image segmentation of an improved CV model.
Detailed Description
The invention is further described with reference to the following drawings and detailed description.
Example 1: as shown in fig. 1, a method for segmenting a level set image based on a curve area includes the following specific steps:
step 1: and drawing a closed curve for any image to be segmented to divide the image into an inner area and an outer area.
Step 2: and calculating the area energy of the whole image region and the area energy inside the closed curve, wherein the area energy outside the closed curve is equal to the area energy of the whole image region minus the area energy inside the closed curve.
Step 3: an image curve area energy constraint term is calculated that is equal to the area energy outside the closed curve minus the area energy inside the closed curve.
Step 4: calculating an overall image energy general function which is equal to the sum of 4 terms; the first term is the length term of the closed curve, the second term is the curve area energy constraint term, and the third and fourth terms are the inner area energy term and the outer area energy term, respectively.
Step 5: and when the closed evolution curve is positioned at the boundary of the target to be segmented, the total energy general function of the image is minimum, and the segmented image is obtained.
In step S2, the process is described by taking the initial calculation parameter Δ t (S) of the lung CT scan image in fig. 2 as 0.1, which is specifically as follows:
(1): calculating the area energy of the whole image region: suArea energy of whole image area
Figure BDA0002321801810000031
Wherein u is0＝[0,195]Representing the gray value of any image to be segmented (in this case, the lung CT scan image), H (phi) is the Heaviside function, phi is the level set function, and the value of the Heaviside function is 0.3183 in this case, and the edge detection function
Figure BDA0002321801810000032
▽ denotes the gradient operator, GσGauss function representing window size sigma
Figure BDA0002321801810000033
The window size is then 5 x 5, Gσ1.5, the value of the gaussian function is Gσ＝[0.0144,0.0853]U is a constant 650.25 and Ω is the image space u0One set of (a).
(2): calculate the area energy inside the closed curve:
Figure BDA0002321801810000034
(3): calculate the area energy outside the closed curve: soutside(C)Is the area energy outside the closed curve,
Soutside＝Su-Sinside＝∫Ωg*u0*H(φ)-g*H(φ)dxdy＝[-0.32,7.78]
the step S3 is specifically as follows:
calculating an image curve area energy constraint term:
Figure BDA0002321801810000035
the step S4 is specifically as follows:
calculating the energy general function of the whole image:
Figure BDA0002321801810000036
Figure BDA0002321801810000037
wherein, the first term of the equation is the length term of the curve C, the coefficient of which the length term is mu ≧ 0 is 650.25, the second term of the equation is the area term fused with the edge information curve C, the coefficient of which the area term is v ≧ 0 is 1, λ1,λ2The coefficient of the energy term of the inner area and the outer area is more than or equal to 0 and is respectively equal to 1 and c1,c2Equal to 19.1266 and 39.0575, respectively, representing the mean gray value inside the evolution curve and the mean gray value outside the evolution curve, δε(φ)＝[0.0187,0.3183]Representing the Dirac function.
The original image in fig. 2 was iteratively divided by the above method, and the results are shown in table 1:
Figure BDA0002321801810000041
TABLE 1
Table 1 shows the image segmentation iteration number and the running time experiment result of fig. 2, and it can be concluded that the improved CV model far exceeds the conventional CV model in terms of running efficiency, but after the CV model is sufficiently iterated, the evolution curve fails to reach the outer contour edge of the target.
While the present invention has been described in detail with reference to the embodiments shown in the drawings, the present invention is not limited to the embodiments, and various changes can be made without departing from the spirit and scope of the present invention.

Claims (4)

1. A level set image segmentation method based on curve area is characterized in that:
step 1: drawing a closed curve on any image to be segmented to divide the image into an inner area and an outer area;
step 2: calculating the area energy of the whole image region and the area energy inside the closed curve, wherein the area energy outside the closed curve is equal to the area energy of the whole image region minus the area energy inside the closed curve;
step 3: calculating an image curve area energy constraint term which is equal to the area energy outside the closed curve minus the area energy inside the closed curve;
step 4: calculating an overall image energy general function which is equal to the sum of 4 terms; the first term is the length term of the closed curve, the second term is the curve area energy constraint term, and the third term and the fourth term are the internal area energy term and the external area energy term respectively;
step 5: and when the closed evolution curve is positioned at the boundary of the target to be segmented, the total energy general function of the image is minimum, and the segmented image is obtained.
2. The curve area-based level set image segmentation method as set forth in claim 1, wherein Step2 is specifically:
(1): calculating the area energy S of the whole image regionu，
Figure FDA0002321801800000011
Wherein u is0Representing the gray value of any image to be segmented, wherein H (phi) is a Heaviside function, phi is a level set function, and phi is an edge detection function
Figure FDA0002321801800000012
▽ denotes the gradient operator, GσGaussian function representing window size σ
Figure FDA0002321801800000013
u is a constant and Ω is the image space u0One set of (2);
(2): calculating the area energy S inside the closed curveinside(C)，
Figure FDA0002321801800000014
(3): calculating the area energy S outside the closed curveoutside(C)，
Soutside＝Su-Sinside＝∫Ωg*u0*H(φ)-g*H(φ)dxdy。
3. The curve area-based level set image segmentation method as set forth in claim 1, wherein Step3 is specifically:
calculating an image curve area energy constraint term Econv，
Figure FDA0002321801800000015
4. The curve area-based level set image segmentation method as set forth in claim 1, wherein Step4 is specifically:
calculating the energy general function of the whole image:
Figure FDA0002321801800000016
Figure FDA0002321801800000021
wherein, the first term of the equation is the length term of the curve C, mu is more than or equal to 0 and is the coefficient of the length term, the second term of the equation is the area term of the curve C with the edge information fused, u is the area term of the curve C0Representing the gray value of any image to be segmented, v is more than or equal to 0 and is the coefficient of area term, lambda1,λ2A coefficient of energy terms in inner and outer regions, c1,c2Respectively representing the mean grey value inside the evolution curve and the mean grey value outside the evolution curve, deltaεAnd (phi) represents the Dirac function.
CN201911301075.2A 2019-12-17 2019-12-17 Level set image segmentation method based on curve area Pending CN111127479A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201911301075.2A CN111127479A (en) 2019-12-17 2019-12-17 Level set image segmentation method based on curve area

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201911301075.2A CN111127479A (en) 2019-12-17 2019-12-17 Level set image segmentation method based on curve area

Publications (1)

Publication Number Publication Date
CN111127479A true CN111127479A (en) 2020-05-08

Family

ID=70498390

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201911301075.2A Pending CN111127479A (en) 2019-12-17 2019-12-17 Level set image segmentation method based on curve area

Country Status (1)

Country Link
CN (1) CN111127479A (en)

Citations (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20100322521A1 (en) * 2009-06-22 2010-12-23 Technion Research & Development Foundation Ltd. Automated collage formation from photographic images
CN102354396A (en) * 2011-09-23 2012-02-15 清华大学深圳研究生院 Method for segmenting image with non-uniform gray scale based on level set function
CN102592287A (en) * 2011-12-31 2012-07-18 浙江大学 Convex optimization method for three-dimensional (3D)-video-based time-space domain motion segmentation and estimation model
CN103295218A (en) * 2012-03-02 2013-09-11 华为技术有限公司 Image cutting method and device
CN104574430A (en) * 2015-02-09 2015-04-29 重庆大学 CT image crack segmentation method based on C-V and RSF models
CN105869178A (en) * 2016-04-26 2016-08-17 昆明理工大学 Method for unsupervised segmentation of complex targets from dynamic scene based on multi-scale combination feature convex optimization
CN107016683A (en) * 2017-04-07 2017-08-04 衢州学院 The level set hippocampus image partition method initialized based on region growing
CN107727662A (en) * 2017-09-28 2018-02-23 河北工业大学 A kind of cell piece EL black patch detection methods based on algorithm of region growing
CN108090909A (en) * 2017-12-15 2018-05-29 中国人民解放军陆军军医大学第附属医院 A kind of ultrasonic contrast image partition method based on statistics Partial Differential Equation Model
CN109087309A (en) * 2018-07-19 2018-12-25 华南理工大学 A kind of image partition method of amalgamation of global and local information level collection
CN109598740A (en) * 2018-12-25 2019-04-09 辽宁师范大学 Image partition method based on Dual Action skeleton pattern

Patent Citations (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20100322521A1 (en) * 2009-06-22 2010-12-23 Technion Research & Development Foundation Ltd. Automated collage formation from photographic images
CN102354396A (en) * 2011-09-23 2012-02-15 清华大学深圳研究生院 Method for segmenting image with non-uniform gray scale based on level set function
CN102592287A (en) * 2011-12-31 2012-07-18 浙江大学 Convex optimization method for three-dimensional (3D)-video-based time-space domain motion segmentation and estimation model
CN103295218A (en) * 2012-03-02 2013-09-11 华为技术有限公司 Image cutting method and device
CN104574430A (en) * 2015-02-09 2015-04-29 重庆大学 CT image crack segmentation method based on C-V and RSF models
CN105869178A (en) * 2016-04-26 2016-08-17 昆明理工大学 Method for unsupervised segmentation of complex targets from dynamic scene based on multi-scale combination feature convex optimization
CN107016683A (en) * 2017-04-07 2017-08-04 衢州学院 The level set hippocampus image partition method initialized based on region growing
CN107727662A (en) * 2017-09-28 2018-02-23 河北工业大学 A kind of cell piece EL black patch detection methods based on algorithm of region growing
CN108090909A (en) * 2017-12-15 2018-05-29 中国人民解放军陆军军医大学第附属医院 A kind of ultrasonic contrast image partition method based on statistics Partial Differential Equation Model
CN109087309A (en) * 2018-07-19 2018-12-25 华南理工大学 A kind of image partition method of amalgamation of global and local information level collection
CN109598740A (en) * 2018-12-25 2019-04-09 辽宁师范大学 Image partition method based on Dual Action skeleton pattern

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
JIANFENG HE. ET AL: "Image segmentation of CV model based on Curve Area Constraint", 《ADVANCES IN INTELLIGENCE SYSTEM AND COMPUTING》, vol. 856, 5 October 2018 (2018-10-05), pages 502 - 509 *

Similar Documents

Publication Publication Date Title
CN109472792B (en) Local energy functional and non-convex regular term image segmentation method combining local entropy
CN103390280A (en) Rapid threshold segmentation method based on gray level-gradient two-dimensional symmetrical Tsallis cross entropy
CN102135606B (en) KNN (K-Nearest Neighbor) sorting algorithm based method for correcting and segmenting grayscale nonuniformity of MR (Magnetic Resonance) image
CN110363775B (en) Image segmentation method based on regional variation level set
CN106340010A (en) Corner detection method based on second-order contour difference
CN110136146B (en) SAR image water area segmentation method based on sinusoidal SPF distribution and level set model
CN107424153B (en) Face segmentation method based on deep learning and level set
CN117173201B (en) Second order differential image segmentation method, system, medium and device
CN111553926B (en) Threshold segmentation method based on two-dimensional Tsallis gray scale entropy fast iteration
CN111127479A (en) Level set image segmentation method based on curve area
CN110517271B (en) Image level set segmentation method based on prior shape constraint
CN110176021B (en) Level set image segmentation method and system for saliency information combined with brightness correction
CN115965642A (en) Image segmentation method and device based on adaptive fractional order edge stop function
CN107330869B (en) Reconstruction method of abnormal pixel points after segmentation of overlapped cells
CN114596320A (en) Image segmentation method and device based on ALSLCV model
CN115619795A (en) Brain MR and natural image segmentation method based on skewed distribution
CN111598890B (en) Level set optimization method for underwater image segmentation
CN115393172A (en) Method and equipment for extracting light stripe centers in real time based on GPU
CN111915632B (en) Machine learning-based method for constructing truth database of lean texture target object
CN107798684A (en) A kind of active contour image partition method and device based on symbol pressure function
CN109685114B (en) Image similarity judgment method based on pre-screening method and PHash
CN112102350A (en) Secondary image segmentation method based on Otsu and Tsallis entropy
Ruzaeva et al. A hybrid multi-object segmentation framework with model-based B-splines for microbial single cell analysis
CN114119622B (en) Active contour image segmentation method and device based on super pixels and sparse constraint
CN112700459A (en) Level set infrared image segmentation method based on multi-feature information fusion

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
AD01 Patent right deemed abandoned

Effective date of abandoning: 20231017

AD01 Patent right deemed abandoned