Proceedings of the IEEK Conference (대한전자공학회:학술대회논문집)

The Institute of Electronics and Information Engineers (대한전자공학회)
권호 표지

Automatic Extraction of the Interest Organization from Full-Color Continuous Images for a Biological Sample

  • Takemoto, Satoko (Department of Information and Computer Science, Saitama University) ;
  • Yokota, Hideo (Department of Information and Computer Science, Saitama University) ;
  • Shimai, Hiroyuki (Department of Information and Computer Science, Saitama University) ;
  • Makinouchi, Akitake (Department of Information and Computer Science, Saitama University) ;
  • Mishima, Taketoshi (Department of Information and Computer Science, Saitama University)
  • Published : 2002.07.01
Download PDF
⟨ Previous Next ⟩

Abstract

We presented the automatic extraction technique of a biological internal organization from full-color continuous images. It was implemented using the localized homogeneousness of color intensity, and also using the continuity between neighboring images. Moreover, we set the "four-level status value" of area condition as a value showing "area possibility. This played important role of preventing a miss-judgement of area definition. These our approach had a beneficial effect on tracking color and shape change of the interest area in continuous extraction. As a resell we succeeded in extraction of mouse's stomach from continuous 50 images.

Keywords