3D Reconstruction Using Multiple Acoustic Images Under Roll Motion Based on Backprojection Techniques

Abstract

In underwater with high levels of turbidity, a multibeam acoustic camera proves to be advantageous for obtaining visibility. Unfortunately, the loss of elevation information during the making process of 2D acoustic images causes it challenging to accurately determine the three-dimensional information of objects. This paper proposes a method that applies the Radon transform and back-projection method to restore elevation and reconstruct objects in 3D from 2D acoustic images. For this, 2D acoustic images need to be taken at different perspective views as the camera undergoes a rolling motion from 0 to 90 degrees. Experiment results are included as well.