Canny Operator Based Level Set Segmentation Algorithm for Medical Images

Abstract

The segmentation and extraction of tissues and organs are fundamental work in 3D reconstruction and visualization of medical images. Considering the features of virtual human images, a Canny operator based level set algorithm and the analytic expression of Level set equation is deduced. The algorithm is implemented using narrow band method numerically. The algorithm combines the advantages of Canny operator which can orient the boundary accurately and the idea that Level set method continuously evolves the boundary in image space. Experiments show that the algorithm calculates the result precisely with good anti-jamming property and fast computing speed. Fine results can be obtained by using this algorithm.