Region extraction method based on clustering along an object contour

Y. Matsuzawa, I. Kumazawa, T. Abe

Research output: Contribution to conferencePaperpeer-review


For accurate region extraction, region-based active contour models (ACM) have been proposed. Compared with ordinary ACM, they require heavier loads of initial settings and further processing time. Considering these problems, for effective and efficient extraction, we propose a new region-based ACM. In the proposed method, first, users draw an initial curve in an object. Along this curve, definite length scanlines are set perpendicular to the initial curve and across the object contour. Depending on the image properties around scanlines, the scanlines are separated into several groups and each group is divided into two subregions (object and background regions). Scanlines grouping and groups dividing are iterated until the image properties are sufficiently uniform in each subregion. Through this clustering, a region along an object contour is segmented into subregions of uniform image properties, and consequently the image properties in a necessary and sufficient area are introduced into extraction process effectively.

Original languageEnglish
Number of pages4
Publication statusPublished - 2001 Jan 1
Externally publishedYes
EventIEEE International Conference on Image Processing (ICIP) - Thessaloniki, Greece
Duration: 2001 Oct 72001 Oct 10


OtherIEEE International Conference on Image Processing (ICIP)

ASJC Scopus subject areas

  • Hardware and Architecture
  • Computer Vision and Pattern Recognition
  • Electrical and Electronic Engineering


Dive into the research topics of 'Region extraction method based on clustering along an object contour'. Together they form a unique fingerprint.

Cite this