Title :
Use of a JPEG-2000 Wavelet Compression Scheme for Content-Based Ophtalmologic Retinal Images Retrieval.
Author :
Lamard, Mathieu ; Daccache, Wissam ; Cazuguel, Guy ; Roux, Christian ; Cochener, Béatrice
Author_Institution :
Laboratoire de Traitement de l´´Information Medicale, INSERM, Brest
Abstract :
In this paper we propose a content based image retrieval method for diagnosis aid in diabetic retinopathy. We characterize images without extracting significant features, and use histograms obtained from the compressed images in JPEG-2000 wavelet scheme to build signatures. The research is carried out by calculating signature distances between the query and database images. A weighted distance between histograms is used. Retrieval efficiency is given for different standard types of JPEG-2000 wavelets, and for different values of histogram weights. A classified diabetic retinopathy image database is built allowing algorithms tests. On this image database, results are promising: the retrieval efficiency is higher than 70% for some lesion types
Keywords :
biomedical optical imaging; content-based retrieval; diseases; eye; image classification; image coding; image retrieval; medical image processing; wavelet transforms; JPEG-2000 wavelet compression; classified diabetic retinopathy image database; content-based ophthalmologic retinal images retrieval; database images; diabetic retinopathy diagnosis; query images; weighted distance; Content based retrieval; Diabetes; Feature extraction; Histograms; Image coding; Image databases; Image retrieval; Retina; Retinopathy; Spatial databases;
Conference_Titel :
Engineering in Medicine and Biology Society, 2005. IEEE-EMBS 2005. 27th Annual International Conference of the
Conference_Location :
Shanghai
Print_ISBN :
0-7803-8741-4
DOI :
10.1109/IEMBS.2005.1615341
