Image retrieval based on bit-plane entropy and spatial entropy
Xiaosheng Wu · Systems engineering and electronics · 2009
For the image retrieval method based on bit-plane entropy(BE),the main problems are that it is easily affected by the symmetry of entropy and does not take the bit-plane spatial distribution feature into consideration.In order to remove the effect of these two drawbacks,a novel image retrieval algorithm is proposed.Firstly,enhanced bit-plane entropy(EBE) is introduced to avoid the influence of the symmetry of entropy.Then,bit-plane spatial entropy(SE) is presented to describe the spatial feature of a bit-plane.Based on EBE and SE,the similarity of images is discussed in detail and three measures are given.In comparison with bit-plane distribution entropy,experiment results show that the proposed method gives better performance.