A fast geometric active contour model with automatic region grid
Krishnan Nallaperumal, S. Naga Nandini Sujatha · 2010
A new frame work for accurate extraction of object boundary from back ground for cell nucleus segmentation of cervical cancer images is proposed in this work. It proposes a new region grid algorithm for the fast evolution of level-set-based geometric active contours and compares it with other established numerical schemes. Geometric active contour models are very popular partial differential equation-based tools in object boundary extraction. This algorithm overcomes the drawback associated with the implementation of implicit schemes. The proposed system is more accurate compared with alternate split schemes. The region grid is formed by image gradient magnitude and directional information. The combined algorithm allows for the rapid evolution of the contour and convergence to its final configuration after very little iteration. The experiments demonstrate the efficiency and accuracy of the method.