Cluster coding with modified flood fill algorithm for texture segmentation
Loading...
Date
2009
Authors
Khoo, Hee Kooi
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Texture refers to properties that represent the surface or structure of an object and is defined as something consisting of mutually related elements. The main focus in this study is to do texture segmentation in two dimensional (2D) digital images. A series of hybrid algorithms to segment textures is proposed in order to maintain or improve the classification accuracy and computation time. The algorithms are divided into four stages; these are feature extraction, feature enhancement, feature selection, and feature smoothening. Grey level co-occurrence probabilities (GLCP) method is being used to extract features from texture images. Statistical features can be calculated based on the GLCP generated. The features are obtained by using a combination of different angles to give some rotational invariance. To enhance the features, histogram equalization (HE) is applied to the statistical features. Cluster coding with modified flood fill algorithm is proposed for feature selection to resolve the uncertain texture patterns, noise, and outliers occurring on the extracted feature domain. The whole texture segmentation system is written in C++ programming language with one dimensional (1D) array data structure for fast computation, especially when dealing with algorithms involving iterations. Brodatz texture album is used in this study to test out the result. In this study, the cluster coding with modified flood fill algorithm showed a significant improvement over other techniques in terms of classification accuracy and computation time.
Description
Master
Keywords
Mathematical science , Cluster coding , Flood fill algorithm