Local and Global Active Contour Model for Image Segmentation With Intensity Inhomogeneity

Cai, Qing; Liu, Huiying; Qian, Yiming; Li, Jing; Duan, Xiaojun; Yang, Yee‐Hong

doi:10.1109/access.2018.2871846

Cited by 13 publications

(4 citation statements)

References 65 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Arnold [59] notes that the function φ, which meets |∇ϕ|= 1 , is an SDF plus a constant [60]. According to the above analysis, Li et al [61] designed an algorithm without reinitialization.…”

Section: Distance Regularization Termmentioning

confidence: 99%

An Image Segmentation Method Based on Improved Regularized Level Set Model

Sun

Meng

et al. 2018

Applied Sciences

View full text Add to dashboard Cite

When the level set algorithm is used to segment an image, the level set function must be initialized periodically to ensure that it remains a signed distance function (SDF). To avoid this defect, an improved regularized level set method-based image segmentation approach is presented. First, a new potential function is defined and introduced to reconstruct a new distance regularization term to solve this issue of periodically initializing the level set function. Second, by combining the distance regularization term with the internal and external energy terms, a new energy functional is developed. Then, the process of the new energy functional evolution is derived by using the calculus of variations and the steepest descent approach, and a partial differential equation is designed. Finally, an improved regularized level set-based image segmentation (IRLS-IS) method is proposed. Numerical experimental results demonstrate that the IRLS-IS method is not only effective and robust to segment noise and intensity-inhomogeneous images but can also analyze complex medical images well.

show abstract

Section: Distance Regularization Termmentioning

confidence: 99%

An Image Segmentation Method Based on Improved Regularized Level Set Model

Sun

Meng

et al. 2018

Applied Sciences

View full text Add to dashboard Cite

show abstract

“…The Mumford-Shah (M-S) [1] model-based C-V model employs the global intensity difference between the inner and outer regions of the contour's average intensities to drive the segmentation process is a prominent region-based active contour model. This region-based method is less susceptible to initial contour placement and works better with images with hazy edge points [5], [44]. Despite advances in both edge-based and region-based active contour models, researchers have discovered some drawbacks that restrict their effectiveness in real-world applications [47], [48].…”

Section: Introductionmentioning

confidence: 99%

Image Segmentation Using Bias Correction Active Contours

Zia,

Soomro,

Choi

2024

IEEE Access

View full text Add to dashboard Cite

Deep learning-based image segmentation methods require densely annotated and massive datasets to produce effective results. On the other hand, active contours-based methods are excellent alternatives to the situation, producing acceptable segmentation results. Earlier active contour models, including local and global region information, struggle with their limitations, such as spurious contours appearing in inhomogeneous images. Bias correction is utilized to solve the bias field's energy, considering the intensity inhomogeneity and the level set functions that suggest an image domain division. In our approach, we combine the advantages of local and global information in the image level set function, resulting in a combined energy function that aids in the efficient evolution of contours on images and can judge the relevance of the item and its surroundings. The proposed model computes data force by extracting local information from an in-homogeneous image using image-fitting energy and then computing all pixel values simultaneously. Objects with high differences between grey levels or more in-homogeneity can be segmented. The outcome demonstrates that our method is more dependable and computationally efficient than previous methods.

show abstract

“…This method can be divided into parametric active contour model (PACM) and geometric active contour model (GACM). The earliest PACM, the Snake model, was proposed by Kass et al [12], which is limited by a precise location of the initial contour curve and cannot segment edge depression images [13]. It is well known that the PACM generally has the following common problem: it is unable to deal with changes in the topological structure of the curve (such as curve splitting or merging).…”

Section: Introductionmentioning

confidence: 99%

Active contour model based on hybrid signed pressure force function

Meng

2023

IET Image Processing

View full text Add to dashboard Cite

To segment noisy and multi‐target images, an active contour model based on a hybrid signed pressure force function that fuses global and local information of the image is proposed. Firstly, a local signed pressure force function is defined using the local area information of the image. Then, it is combined with the existing global grey density function to construct a new signed pressure force function. Finally, the selective binary and Gaussian filtering regularized level set are modified using the newly defined function. Extensive experiments and comparisons on both synthetic and real images demonstrate that the proposed method is robust to noise and can handle noisy and multi‐target images with high accuracy.

show abstract

Local and Global Active Contour Model for Image Segmentation With Intensity Inhomogeneity

Cited by 13 publications

References 65 publications

An Image Segmentation Method Based on Improved Regularized Level Set Model

An Image Segmentation Method Based on Improved Regularized Level Set Model

Image Segmentation Using Bias Correction Active Contours

Active contour model based on hybrid signed pressure force function

Contact Info

Product

Resources

About