Share Email Print

Proceedings Paper

Segmentation of pomegranate MR images using spatial fuzzy c-means (SFCM) algorithm
Author(s): Ghobad Moradi; Mousa Shamsi; M. H. Sedaaghi; M. R. Alsharif
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

Segmentation is one of the fundamental issues of image processing and machine vision. It plays a prominent role in a variety of image processing applications. In this paper, one of the most important applications of image processing in MRI segmentation of pomegranate is explored. Pomegranate is a fruit with pharmacological properties such as being anti-viral and anti-cancer. Having a high quality product in hand would be critical factor in its marketing. The internal quality of the product is comprehensively important in the sorting process. The determination of qualitative features cannot be manually made. Therefore, the segmentation of the internal structures of the fruit needs to be performed as accurately as possible in presence of noise. Fuzzy c-means (FCM) algorithm is noise-sensitive and pixels with noise are classified inversely. As a solution, in this paper, the spatial FCM algorithm in pomegranate MR images' segmentation is proposed. The algorithm is performed with setting the spatial neighborhood information in FCM and modification of fuzzy membership function for each class. The segmentation algorithm results on the original and the corrupted Pomegranate MR images by Gaussian, Salt Pepper and Speckle noises show that the SFCM algorithm operates much more significantly than FCM algorithm. Also, after diverse steps of qualitative and quantitative analysis, we have concluded that the SFCM algorithm with 5×5 window size is better than the other windows.

Paper Details

Date Published: 30 September 2011
PDF: 7 pages
Proc. SPIE 8285, International Conference on Graphic and Image Processing (ICGIP 2011), 82850N (30 September 2011); doi: 10.1117/12.913289
Show Author Affiliations
Ghobad Moradi, Sahand Univ. of Technology (Iran, Islamic Republic of)
Mousa Shamsi, Sahand Univ. of Technology (Iran, Islamic Republic of)
M. H. Sedaaghi, Sahand Univ. of Technology (Iran, Islamic Republic of)
M. R. Alsharif, Univ. of the Ryukyus (Japan)

Published in SPIE Proceedings Vol. 8285:
International Conference on Graphic and Image Processing (ICGIP 2011)
Yi Xie; Yanjun Zheng, Editor(s)

© SPIE. Terms of Use
Back to Top