Cellular automaton based pixel-level snakes
Tomohiro Fujita, Syo Sawada, Kaihei Iwanaga, Takeshi Kumaki, Mamoru Nakanishi, Takeshi Ogura · 2014
A new version of Pixel Level Snake (PLS), which is an image segmentation algorithm, is proposed. The proposed algorithm is based on processes of Cellular Automaton (CA). In CA, the forces of resistance to stretch and bending are defined by the CA rule, and comparing these forces and the gradients of image pixels determines deforming of the active contour. The CA based PLS consists of simple transition rules, and therefore easy implementation and high speed processing are expected on our dedicated hardware engine. In our experiments, we confirmed the feasibility of the image segmentation with the CA based PLS.