Biological image segmentation based on object pigmentation

Author

Zheng, Jue ; Valavanis, Kimon ; Gauch, John

Author_Institution

Northeastern Univ., Boston, MA, USA

fYear

1991

fDate

4-5 Apr 1991

Firstpage

204

Lastpage

205

Abstract

A methodology for segmenting biological images based on object pigmentation is proposed. The object of interest is extracted based on MacAdam´s just noticeable difference (see J. Opti. Soc. Am., vol.32, p.247-74, 1942). The method combines pictorial and spectral information, into a single valued total color difference. It has two important properties: accuracy based on human color observation and easy implementation. Experiments show that the proposed method is relatively insensitive to nonuniform illumination. The authors briefly review the MacAdam´s ellipses in colorimetry and derive a total color difference based on the MacAdam´s ellipses

Keywords

biological techniques and instruments; colour; picture processing; MacAdam´s ellipses; MacAdam´s just noticeable difference; biological image segmentation; colorimetry; human color observation; nonuniform illumination; object pigmentation; pictorial information; spectral information; total color difference; Biology computing; Biomedical imaging; Biomedical measurements; Color; Computer science; Computerized monitoring; Equations; Image segmentation; Pigmentation; Sea measurements;

fLanguage

English

Publisher

ieee

Conference_Titel

Bioengineering Conference, 1991., Proceedings of the 1991 IEEE Seventeenth Annual Northeast

Conference_Location

Hartford, CT

Print_ISBN

0-7803-0030-0

Type

conf

DOI

10.1109/NEBC.1991.154646

Filename

154646