PROJECT TITLE :

Vision-based cleaning area control for cleaning robots

ABSTRACT :

This paper provides a vision primarily based HCI technique for a user to command a cleaning robot to maneuver to a selected location in home environment. Six hand poses are detected from a video sequence taken from a camera on the cleaning robot. AdaBoost based mostly hand-pose detectors are trained with a reduced Haar-like feature set to form the detectors sturdy to the influence of the advanced background. The primary 3 stages of the cascade in the six detectors are used as Pose Estimation to reduce the computational complexity. The cleaning area is determined from the detected pose. The performances of the proposed detectors are validated with a set of check images with cluttered background. The cleaning space management is simulated with real-world video sequences. The proposed methodology can effectively control a cleaning robot without the necessity for a user to wear or use any input devices.