Object detection and tracking is an important function that helps bringing robots into real-world environments. Detection of an object-of-interest (OOI) from multiple stacking objects still holds certain degree of difficulty. In this paper, a novel target selection and tracking algorithm is proposed for detecting and tracking one OOI from multiple stacking objects. The proposed method first employs a keypoint extraction algorithm to compute control points from incoming images. Next, each OOI appeared in the image is extracted from the multiple stacking objects in a box by using mean shift clustering approach. Finally, a template-based visual tracking method is adopted to locate and track center position of the top OOI in the box. The proposed algorithm had been implemented on an Intel Core i5-4440 3.1GHz platform, achieving real-time performance about 30 frames per second at 640x480 image resolution in the experiments.
Relation:
Proceedings of the 3rd IIAE International Conference on Intelligent Systems and Image Processing 2015, pp.187-190