Shallow Depth SIFT Based Approach for Mapping underwater surfaces using AUV's

Autonomous Underwater Vehicles(AUVs) have always been used in oceanic exploration. They were used for topographic mapping, studying the various elements of the sea from flora and fauna to synthetic objects and debris alike. This paper aims to produce perceivable maps of the shallow seabed using AUVs, using the Scale-Invariant Feature Transform(SIFT) algorithm for collecting features from the images and using the Brute-force matcher to match the images producing 2-dimensional rectangular maps by applying the proposed algorithm. The algorithm was tested using handheld webcams in a simulated environment and our results were consistent with the expected output, and also account for the uncertainty of noise, distortion due to the reflection of light on the surface of the water at shallow depths. This approach to mapping has a low computational cost and can be deployed to multiple AUVs to map larger areas.