Multi-beam Sonar Image Sequences

The dataset supports the following article: Diamantis K., Yaghoobi M., Davies M. and Hopgood J. 2019, 'A multi-frame resolution enhancement framework for multi-beam sonar systems' submitted for review to the IEEE Transactions of Geoscience and Remote Sensing. The article proposes a framework for the improvement of multi-beam echo-sounder image quality with the objective to facilitate real-time high-resolution applications. The framework relies on the fact that a large number of images are commonly acquired from different positions, during underwater scans. The dataset consists of seven different multi-beam sonar image sequences, used in the studies of improving sonar image resolution. The image sequences depict various underwater objects/structures and are converted to video-clips. Each video-clip includes from left to right: the standard sonar images, the images resulting from image registration and averaging, and the reconstructed images using the proposed high-resolution framework. All images have been processed using Matlab (The MathWorks, Inc., Natick, MA, USA) scripts.