Abstract:

Image retrieval in discrete cosine transform (DCT) compressed domain is proposed by introducing the spatial information and the energy of DCT blocks into the feature extraction. According to the distribution of DCT coefficients, the spatial distribution map is constructed and spatial distribution feature is given. Meanwhile, the effect of the energy of DCT blocks on retrieval precision is considered, an energy distribution feature is presented as another feature. The energy distribution of DCT blocks and the spatial distribution of DCT coefficients are taken into account for avoiding the retrieval mistaken and lose. The experiment results show that the new method has a good performance both in retrieval efficiency and effectiveness.

Key words: content-based image retrieval, discrete cosine transform , compressed domain, spatial distribution map