Top-Down Segmentation of Histological Images Using a Digital Deformable Model

Abstract. This paper presents a straightforward top-down segmentation method based on a contour approach on histological images. Our approach relies on a digital deformable model whose internal energy is based on the minimum length polygon and that uses a greedy algorithm to minimise its energy. Experiments on real histological images of breast cancer yields results as good as that of classical active contours.

• Type: Conference
• Year: 2009
• Where: ISVC
• Authors: François de Vieilleville, Jacques-Olivier Lachaud, Paulette Herlin, Olivier Lezoray, Benoît Plancoulaine
• Keywords: Applied Computing, Histological Images, ISVC 2009, Real Histological Images, Straightforward Top-down Segmentation