Integrating Kinect Depth Data with a Stochastic Object Classification Framework for Forestry Robots

Abstract

Abstract: In this paper we study the integration of a depth sensor and an RGB camera for a stochastic classification system for forestry robots. The images are classified as bush, tree, stone and human and are expected to come from a robot working in forest environment. A set of features is extracted from labeled images to train a number of stochastic classifiers. The outputs of the classifiers are then combined in a meta-classifier to produce the final result. The results show that using depth information in addition to the RGB results in higher classification performance. 1