Distance measure with computer vision and neural networks for underwater applications

This document describes a distance measurement system for underwater applications with computer vision and an Artificial Neural Network (ANN). The developed system is installed in the observation class remotely operated vehicle (ROV) VISOR3, to be used in experimental testing. An integrated camera with two laser pointers are used to obtain input data to a Multi-Layer Perceptron (MLP) ANN, trained with a resilient backpropagation algorithm (RPROP). The solution described in this paper is an alternative to other methods found in the literature. Experimental results are presented.