Aspects of image compression using neural networks for visual servoing in robot control

Abstract

Artificial intelligence is widely used in image processing. Neural networks (NN) were successful used for solving complicated issues due to their capacity of generalization and learning from examples. In this paper some aspects of image compression using artificial neural networks are discussed. The network is used in the feedback loop of the visual servoing system, which aims to control a wheeled mobile robot equipped with a robotic manipulator. Hence, grayscale images are used. The goal is to find a low-complexity feed-forward neural network (FFNN) model for image compression, with good compression rate, but with small errors between original image and the reconstructed one. The model is validated using the invariant moments of the image. Different FFNN architectures, with different data sets and training conditions, are analyzed.