Shortest Path Problem with Pixel Level Snakes: Application to Robot Path Planning

摘要

ABSTRACT: In this paper the application of the CNN-based active contour technique called pixel-level snakes (PLS) to find the shortest path in planar labyrinths is proposed. The strategy relies on multiple processing steps with a PLS algorithm implemented on the CNNUM. The proposal has been extended to approximate the shortest length path in sparse labyrinths suitable for applications like robot path planning. 1.