Robot vision - Color based human tracking using the UV model

Abstract

The main objective of robotic research is to aid industrialism, making human lives convenient and flexible with the use of robots. This work addresses a scenario where a mobile robot tracks and follows a human. Some of the conventional methods such as Neural network, color skin probability map, Heuristic rule and Bayesian decision theory are used in this work. The real-time face detection module, based on a two-dimensional color model in the YUV color space is selected over the traditional skin color model in a three-dimensional color space. A modified CAMSHIFT tracking mechanism in a one-dimensional HSV color space is developed and implemented onto the mobile robot. In addition to the visual cues, the tracking process considers sixteen sonar-scan and tactile sensor readings to generate a robust measure of the person's distance from the robot. The proposed approach is orientation invariant under varying lighting conditions and under natural transformations.