A.Tanuja Devi
Jawaharlal Nehru Technological University, Vijayanagaram, India

The extensive work in SONAR is oceanic Engineering which is one of the most developing researches in engineering. The SideScan Sonars (SSS) are one of the most utilized devices to obtain acoustic images of the seafloor. This paper proposes an approach for developing an efficient system for automatic object detection utilizing the technique of accumulated cell average-constant false alarm rate in 2D (ACA-CFAR-2D), where the optimization of the computational effort is achieved. This approach employs image segmentation as preprocessing step for object detection, which have provided similar results with other approaches like undecimated discrete wavelet transform (UDWT), watershed and active contour techniques. The SSS sea bottom images are segmented for the 2D object detection using these four techniques and the segmented images are compared along with the experimental results of the proportionof segmented image (P) and runtime in seconds (T) are presented.

Accumulated cell average- constant false alarm rate (ACA-CFAR), two-dimensional object detection, sidescan sonar (SSS), Active contour, Watershed and undecimated discrete wavelet transform.