DocumentCode
2355778
Title
A recursive Otsu-Iris filter technique for high-speed detection of lumen region from endoscopic images
Author
Tian, H. ; Srikanthan, Thambipillai ; Asari, K. Vijayan
Author_Institution
Center for High Performance Embedded Syst., Nanyang Technol. Univ., Singapore
fYear
2001
fDate
1-12 Oct 2001
Firstpage
182
Lastpage
186
Abstract
In this paper, a hardware efficient technique to segment the lumen region from endoscopic images is presented. It is based on applying the combined Otsu-Iris filter operations recursively. The proposed technique applies the Otsu´s procedure recursively to obtain a coarse region of interest (ROI), which is then subjected to an Iris filter operation so that a smaller enhanced region can be identified. This enhanced region is subjected to the Otsu´s procedure recursively and the process of performing Iris filter operation is repeated as before. It has been shown that by repeating this Otsu-Iris filter combination in an iterative manner facilitates in the rapid identification of the lumen region accurately. It has been shown that the proposed method substantially reduces the number of computations of Otsu´s procedure when compared with the APT-Iris filter method. Finally, unlike the APT-Iris filter method, it does not require the precomputation of the cumulative limiting factor, which is highly dependent on the complex endoscopic images
Keywords
image segmentation; medical image processing; recursive filters; APT-Iris filter method; Iris filter operation; coarse region of interest; complex endoscopic images; endoscopic images; high-speed detection; lumen region segmentation; recursive Otsu-Iris filter technique; Cameras; Colon; Data mining; Embedded computing; Embedded system; Endoscopes; Filters; Navigation; Rough surfaces; Surface roughness;
fLanguage
English
Publisher
ieee
Conference_Titel
Applied Imagery Pattern Recognition Workshop, AIPR 2001 30th
Conference_Location
Washington, DC
Print_ISBN
0-7695-1245-3
Type
conf
DOI
10.1109/AIPR.2001.991223
Filename
991223
Link To Document