Publication Details
Weiping Liu, Yanfeng Shang, Xin Yang, Rudi Deklerck, Jan Cornelis

IFMBE Proceedings

Contribution To Book Anthology


An effective shape deformation method derived from a PCA-based statistical shape model (SSM) using the Golden Section Search (GSS) method is presented. The PCA-based SSM has proved to be a simple and effective method in model based segmentation and has been used in medical image analysis for soft tissue extraction. The gradient descent method is widely applied to search the optimal values for the transformation and deformation parameters to fit the model to specific image features. We use Golden Section Search instead, yielding some conceptual advantages, and experimentally illustrated good performance as well as lower execution times compared to gradient descent methods.