Title :
A fast retrieval of texture images coded with lifting schemes
Author :
Benazza-Benyahia, A. ; Souidène, W.
Author_Institution :
Departement de Mathematiques Appliquees, Signal et Commun., Ecole Superieure des Commun. de Tunis, Ariana, Tunisia
Abstract :
We are interested in indexing images in the compressed domain without decoding these images. More precisely, we focus on applications requiring lossless and progressive reconstruction of images such as telebrowsing of medical or remote sensed images. For this purpose, the image is firstly decomposed with a lifting scheme which provides an exact and compact multiresolution representation of the input image. Then, the salient features are extracted from the coded coefficients. We propose a fast retrieval procedure thanks to an appropriate organization of the features space. Experimental results indicate good performances in term of search time.
Keywords :
data compression; database indexing; image coding; image reconstruction; image representation; image retrieval; image texture; visual databases; experimental results; feature extraction; image coding; image compression; image indexing; image reconstruction; lifting schemes; medical images; multiresolution representation; remote sensed images; search time; telebrowsing; texture image retrieval; Biomedical imaging; Content based retrieval; Decoding; Feature extraction; Image coding; Image resolution; Image retrieval; Image storage; Indexing; Space technology;
Conference_Titel :
Systems, Man and Cybernetics, 2002 IEEE International Conference on
Print_ISBN :
0-7803-7437-1
DOI :
10.1109/ICSMC.2002.1173272