학술논문
Medical images stabilization using sparse-induced similarity measure
Document Type
Conference
Source
2015 2nd International Conference on Pattern Recognition and Image Analysis (IPRIA) Pattern Recognition and Image Analysis (IPRIA), 2015 2nd International Conference on. :1-4 Mar, 2015
Subject
Language
Abstract
Medical image stabilization has been widely used for clinical imaging modalities. Registration is a conspicuous stage for stabilizing dynamic medical images. Some of regular methods are sensitive to bias field distortion. Sparse-induced similarity measure (SISM) is a robust registering method against spatially-varying intensity distortions which is not evitable on clinical imaging instruments. This paper presents a method for registering medical images to average of captured images using SISM method to avoid spatially-varying intensity distortions like Bias field. Proposed method is compared with SSD and MI similarity measure based registrations. Results show enhancement in stabilizing medical dynamic images with SISM method.