文章基本信息

标题：STUDIES ON IMPROVING TEXTURE SEGMENTATION PERFORMANCE USING GENERALIZED GAUSSIAN MIXTURE MODEL INTEGRATING DCT AND LBP
本地全文：下载
作者：K. NAVEEN KUMAR ; K.SRINIVASA RAO ; Y.SRINIVAS 等
期刊名称：Journal of Theoretical and Applied Information Technology
印刷版ISSN：1992-8645
电子版ISSN：1817-3195
出版年度：2016
卷号：92
期号：2
出版社：Journal of Theoretical and Applied
摘要：This paper addresses the performance evaluation of the texture segmentation integrating DCT with LBP. In this method, the whole image is converted in to local binary pattern domain. The LBP image is then divided into different non overlapping blocks. From each block, the DCT coefficients are selected in a zig-zag pattern for each block. Assuming the feature vectors follow a multivariate generalized Gaussian mixture model, the model parameters are estimated using EM algorithm. The initialisation of the model parameters is carried using moment method of estimation and using Hierarchical clustering algorithm. The texture segmentation algorithm is developed under Bayesian frame with component maximum likelihood. The performance of the proposed algorithm is evaluated using performance measures such as GCE, PRI and VOI with randomly selected images from Brodatz database. It is observed that this algorithm outperforms existing texture segmentation algorithms with respect to performance measures.
关键词：Texture Segmentation; Multivariate Generalized Gaussian Mixture Model; Performance Measures; Local Binary Patterns; DCT Coefficients.