Development of image enhancement techniques for object detection in underwater acoustic images

Abstract

Underwater archaeology is the study of undersea environment which involves surveying the sea bed for detection of living and non living resources These types of studies are possible only with sonar systems SONAR Sound Navigation And Ranging technique uses sound as a source for detecting underwater objects by creating acoustic images There are many categories of sonar available among which side scan sonar is one of the cost effective system which is used to create an image map of the sea floor The images captured by side scan sonar are generally noisy low contrast and visually poor quality images The acoustic images are displayed as gray scale images with dark and bright intensities This research work aims at development of new image enhancement algorithms for object detection in underwater acoustic images The image enhancement algorithms involves noise removal or denoising intensity enhancement contrast enhancement edge preserving interpolation for object detection in acoustic imagesThe acoustic images used in this research work were obtained using Edgetech 4125 side scan sonar These images are affected by speckle noise caused by the instrument The speckle noises in the images were removed by the proposed denoising method using Stationary Wavelet Transform and its shrinkage function newline