Color Image Segmentation Using Hybrid
Learning Techniques C. Mala* and M. Sridevi Department of Computer Science and
Engineering National Institute of Technology Tiruchirappalli, India - 620 015 {mala,
msridevi}@nitt.edu
Image segmentation is the
process of finding out all non-overlapping distinct regions from the given image
based on certain criteria such as intensity, color, texture or shape. This
paper proposes a two level hybrid non classical model for image segmentation
based on pixel color and texture features of the image. The first level uses
Fuzzy C-Means (FCM) unsupervised method to form a clustering among all the
pixels based on color and texture properties. And the second level uses
supervised methods [Classification Tree and Adaboost] for color learning and
pixel classification, with the prior knowledge of the image obtained from the
FCM. Experiment was conducted for the three different supervised
classification methods including Support Vector Machine (SVM) and their
performances are analysed. From the results, it is inferred that Adaboost
classifier increases accuracy and reduces misclassification error when
compared to Classification Tree and SVM methods. Keywords: support vector machine, adaboost, classification tree, fuzzy c-mean clustering, texture, color space *Corresponding Author: C. Mala IT
Convergence Practice (INPRA), Vol. 2, No. 2, pp. 21-42, June 2014 [pdf] |