基于位平面熵及分布熵的图像检索

Image retrieval based on bit-plane entropy and spatial entropy

针对基于位平面熵进行图像检索存在的问题,提出了位平面熵的改进方法。该算法采用改进的位平面熵来消除熵的对称性对检索效果造成的影响。提出采用位平面分布熵来描述位平面的空间特征,解决了仅依靠位平面熵检索效率低的问题。结合位平面熵及位平面分布熵,给出了图像间的相似性度量准则。仿真实验证明,该算法无论在检索率、精确度还是平均正确率上都具有更好的检索效果。

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 dis- cussed in detail and three measures are given. In comparison with bit-plane distribution entropy, experiment results show that the proposed method gives better performance.