Author : Deepambika V.A and Arunlal S.L

The three dimensional information about the workspace is indispensable for robot movement and object inspection in robotics and automation applications. Many of these applications rely on stereo correspondence algorithms to acquire the depth information of a scenery. Feature based algorithms gives faster results which involves the extraction of edges and corners. This work proposes a fast stereo matching algorithm that involves image segmentation, centroid computation and disparity mapping. Estimation of an object’s range in this work is done using curve fit.

Keywords: Stereo vision, disparity, feature based algorithm, centroid, curve fit.

Article published in International Journal of Current  Engineering  and Technology, Vol.3,No.5(Dec- 2013)