In this paper, an improved algorithm for the cellular active contour technique called pixel-level snakes is proposed. The motivation is twofold: on the one hand a higher efficiency and flexibility in the contour evolution towards the boundaries of interest is pursued. On the other hand a higher performance and suitability for its hardware implementation onto a CNN chip-set architecture is required. To illustrate the validity of the proposal some examples and data about the computation time from the implementation of the algorithm on the 64/spl times/64 CNNUM chip have been included.