Sonar Image Enhancement based on Minimum Entropy of One-Dimensional Bound Histogram and Rough Sets

Abstract

To enhance a sonar image from a small underwater target, the method using the minimum entropy of the one-dimensional bound histogram and Rough Set Theory is proposed. In the method, the threshold of the gray values of the sonar image is calculated via the minimum value of the entropy of the one-dimensional bound histogram from the sonar image, and the sonar image is divided into some varied sub-images via Rough Set Theory. The varied sub-images are enhanced via their corresponding transforms in gray scale of the sonar image. As an example, an actual sonar image from a small underwater target is enhanced via the proposed method with detailed steps. As a contrast, the sonar image is also enhanced via transformation of histogram equalization. The experimental results show the proposed method is more effective in both improving visualization of the sonar image and preserving its edges.