download0 view776
twitter facebook

공공누리This item is licensed Korea Open Government License

dc.contributor.author
Farhan Akram
dc.contributor.author
최광남
dc.contributor.author
김정헌
dc.contributor.author
이찬건
dc.date.accessioned
2019-08-28T07:41:45Z
dc.date.available
2019-08-28T07:41:45Z
dc.date.issued
2015-01-31
dc.identifier.issn
1748-670X
dc.identifier.uri
https://repository.kisti.re.kr/handle/10580/14438
dc.description.abstract
Segmentation of regions of interest is a well-known problem in image segmentation. This paper presents a region-based image segmentation technique using active contours with signed pressure force (SPF) function. The proposed algorithm contemporaneously traces high intensity or dense regions in an image by evolving the contour inwards. In medical image modalities these high intensity or dense regions refer to tumor, masses, or dense tissues. The proposed method partitions an image into an arbitrary number of subregions and tracks down salient regions step by step. It is implemented by enforcing a new region-based SPF function in a traditional edge-based level set model. It partitions an image into subregions and then discards outer subregion and partitions inner region into two more subregions; this continues iteratively until a stopping condition is fulfilled. A Gaussian kernel is used to regularize the level set function, which not only regularizes it but also removes the need of computationally expensive reinitialization. The proposed segmentation algorithm has been applied to different images in order to demonstrate the accuracy, effectiveness, and robustness of the algorithm
dc.language
eng
dc.relation.ispartofseries
Computational and Mathematical Methods in Medicine
dc.title
Segmentation of Regions of Interest Using Active Contours with SPF Function
dc.citation.endPage
16
dc.citation.startPage
1
dc.citation.volume
2015
dc.subject.keyword
segmentation
dc.subject.keyword
active contours
dc.subject.keyword
signed pressure force
dc.subject.keyword
medical image
Appears in Collections:
7. KISTI 연구성과 > 학술지 발표논문
Files in This Item:
There are no files associated with this item.

Browse