Seabed image texture analysis using subsampled contourlet transform

Abstract

In this paper, a new split and merge algorithm based on the concept of subsampled contourlet transform for automatic segmentation and classification of seafloor images is presented. Contourlet transform is a new extension of the two-dimensional wavelet transform using multiscale and directional filter banks. The subsampled contourlet transform which is introduced by the author is a new version of the contourlet transform that allows analysis of images in square-size coefficients with various resolution levels and directions. It allows analysis of images at various scales as well as directions, which effectively capture smooth contours that are the dominant features in seabed images. The proposed method provides a fast tool with enough accuracy that can be implemented in a parallel structure for real-time processing. In addition, the simulation results are compared with the results of wavelet-based methods as well as other known techniques to show the effectiveness of our proposed algorithm. © 2011 Springer-Verlag.