文章基本信息

标题：Image Segmentation using Bi Directional Self Organize Neural Network (BDSONN)
本地全文：下载
作者：Rudranath Banerjee ; Arun Kanti Manna
期刊名称：International Journal of Computer Science and Information Technologies
电子版ISSN：0975-9646
出版年度：2014
卷号：5
期号：6
页码：7002-7011
出版社：TechScience Publications
摘要：An image segmentation using bi-directional selforganizing neural network (BDSONN) with Multilevel Sigmoidal (MUSIG) function as activation function approach is presented in this Report to segment image in different color levels(for color images) or gray levels(for gray scale images). Multilevel sigmoidal (MUSIG) activation function induces multi scaling capabilities in bi-directional self-organizing neural network (BDSONN) architecture. The function however resorts to equal and fixed class responses, assuming the homogeneity of image information content. The color and gray scale images are segmented by applying the resultant multilevel sigmoidal (MUSIG) activation function. Results of segmentation of color and gray images indicate segmentation efficiency of the BDSONN using MUSIG function
关键词：Image segmentation; Bi-Directional selforganizing;neural network (BDSONN); MUSIG activation;function.