Similarity metric in medical image registration

Autor:	Jing Jin, Yunle Yang
Rok vydání:	2013
Předmět:	Measure (data warehouse) Similarity (network science) business.industry Computer science Metric (mathematics) Image registration Pattern recognition Mutual information Artificial intelligence business Correlation ratio Least squares Field (computer science)
Zdroj:	Proceedings of the 2nd International Conference on Computer Science and Electronics Engineering (ICCSEE 2013).
DOI:	10.2991/iccsee.2013.175
Popis:	Registration algorithm via various similarity measures is presented in this paper. Image registration is a process of align images so that they can get the same structure in the same spatial locations. During the registration, similarity metric is very important to measure the result as a standard. Different similarity metric has its own advantages and application field. Mutual information, correlation ratio and least squares are mentioned in this paper. Problematic issues of these three methods and outlook for the future research are discussed too. The major goal of the paper is to provide a comprehensive comparison among the widely used similarity metric. Keywords-registration; similarity metric; mutual information; correlation ratio; least squares
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_________::90f85901380af19491f6bafc6f5ef1f0 https://doi.org/10.2991/iccsee.2013.175 Zobrazit plný text záznamu