A Novel Adaptive Fractional Differential Active Contour Image Segmentation Method

When the image is affected by strong noise and uneven intensity, traditional active contour models often cannot obtain accurate results. In this paper, a novel adaptive fractional differential segmentation method proposed to solve above problem. At first, in order extract more texture parts of image, an adaptively matrix constructed according gradient information varying each pixel. Then, edge-stopping function regularization term susceptible noise, new fractional-order designed improve resistance. fitting based on differentiation introduced problem improper selection initial position leading inaccurate results so that can be selected arbitrarily. Finally, experimental show effectively accuracy images weak-edge arbitrarily select contour.