Description

We introduce a robust image retrieval system using concentrates on DWT features. Retrieval can be efficiently done using texture features which is obtained using energy distribution over wavelet sub-bands. Wavelet features are obtained by using different types of filters. 2-D discrete wavelet transform is used to extract energy signatures. Searching of relevant images is done by means of computing the euclidean distance of query feature vector with that of feature matrix. After sorting the computed distance values the images those who are closest to the query are considered as retrieved images and they are displayed.