Indexing and retrieval medical images based on 2/spl times/2 DCT and IDS compression
Kamel Saadi, Amira Zemouri, Zahia Brahimi, H. Meraoubi · 2005
Although digital images indexing and querying techniques have extensively been studied for the last years, few systems are dedicated to medical images today while the need for content-based analysis and retrieval tools increases with the growth of digital medical image databases. In this paper, we present a content based medical images indexing and retrieval technique (CBIR) using the 2/spl times/2 discrete cosine transform and information dominance strategy (IDS) compression. The extraction of the feature vectors is based on a 4/spl times/4 DCT model proposed in the literature. These features are directly generated from the 2/spl times/2 DCT coefficients reorganized in subbands and the search process is carried out: by' calculating cosine distance measure between the signatures of the query image and those stored in the database. The developed 2/spl times/2 DCT approach is expected to be very useful for a targeted seeking.