Acoustic Image Enhancement and Segmentation using Multi Support Vector Machine

Underwater Image Processing is challenging due to the physical properties of underwater condition. Poor visibility of underwater images is a noteworthy issue for maritime applications of computer vision. This is mainly due to the presence of haze, fog, scattering and absorption. In order to solve such problem, the proposed technique is used to enhance and restore the underwater images using Multi-SVM (Support Vector Machine) technique. Contrast Limited Adaptive Histogram Equalization (CLAHE) on the RGB image is used to improve the contrast and intensity of the original degraded image. Color correlation technique is carried out for acquiring ideal result on the experimental outcomes both subjectively and objectively. Finally, the underwater images are segmented by the Multi-SVM algorithm. The experimental results demonstrate that the proposed method provide better haze removal than the past strategy and furthermore monitors the underwater scenario effectively through different scenes captured during various climatic conditions. The upgraded images are described by better exposedness of dull areas, enhanced global contrast and edge sharpness. The proposed methodology is used to enhance submerged images with high accuracy and also delivers high quality haze-free images more adequately than the previous method.