Wavelets for content based image retrieval and digital watermarking for multimedia applications

In this paper we have proposed novel wavelet-based texture image retrieval and digital image watermarking techniques. For texture image retrieval we have used standard deviation and energy as a feature measure and Canberra distance metric as dissimilarity measure. Large texture database of 1856 images is used to check the retrieval performance. Proposed retrieval method is far superior to the traditional method, which uses wavelet coefficients energy as a feature and Euclidean distance metric as dissimilarity measure. We have proposed two watermarking algorithms. In first proposed method a binary seal is embedded into the region of interest. This method is robust against lossy compression, distortion due to noise and cropping. In second algorithm spread spectrum technique is used to embed the watermark. This method is robust against median filtering and scaling.