Color Image Segmentation Based on Log Normal Distribution

In this article, we present a novel approach for effective segmentation of color images, by integrating the advantages of the YIQ and the Log Normal Distribution features. Log Normal Distribution is considered in this paper, for effective segmentation, by considering the pattern of the pixels. The proposed method entails less complexity and is thus can be used reasonably very well for realistic image segmentation purposes. The relationship between Hue and Saturation plays a leading role while segmenting the color images which can be observed through the experimental results. The performance evaluation of the developed model is carried out by using metrics like Peak Signal to Noise Ratio (PSNR), Mean Squared Error (MSE), Image Fidelity (IF)

Author(s): Ch. Murali Krishna, Y. Srinivas

Abstract | PDF

Share This Article