Image Contour Extraction Based on CNN and Active Contour Model
Xiaohua Liu, Yuan Da, Jinjiang Li · 2009
A CNN is a dynamic nonlinear system with the local connectivity of cells and easy to be translated into a VLSI implementation. CNNs are very suitable for modeling the physical process of energy propagation since CNNs conserve the physical properties of a continuous structure. For the contour extraction, this paper proposes a method for establishing the evolution model of active contour by CNN technique. Further,the relationship between CNNs and the minimization for active contour model is investigated and the derivation of the CNN implementations for converting the active contour model to CNN templates is performed in order to improve the computing efficiency. Experimental results are given to demonstrate the feasibility of the proposed method in extracting contour from the blurred edge and high-noise images.