作者: Ivan Kitanovski , Katarina Trojacanec , Ivica Dimitrovski , Suzana Loskovska
DOI: 10.1007/978-3-642-37169-1_8
关键词:
摘要: Medical image retrieval is one of the crucial tasks in everyday medical practices. This paper investigates three forms retrieval: text, visual and multimodal retrieval. We investigate by evaluating different weighting models for text In case retrieval, we focused on extracting low-level features examining their performance. For, used late fusion to combine best results. found that choice model dramatically influences outcome The results from are fused using linear combination, which among simplest most frequently methods. Our clearly show with an appropriate technique improves