一种基于位平面综合特征的彩色图像检索方案

An Efficient Color Image Retrieval Technique Based on Multi-Features of Bit-Plane

传统的基于颜色直方图的彩色图像检索方法存在严重不足.首先是丢失颜色空间分布信息及特征维数过高,更重要的是无法有效检索含噪声图像.为克服此缺陷,提出了一种基于位平面综合特征的彩色图像检索算法.首先,结合光照、锐化、模糊等噪声攻击特点,从原始彩色图像中提取出重要位平面;然后选取重要位平面的加权颜色直方图作为颜色特征,选取重要位平面的空间信息熵作为空间特征;再综合利用上述颜色、空间两个特征计算图像间内容的相似度,并进行彩色图像检索.仿真实验表明,算法能够准确和高效地查找出用户所需内容的彩色图像,并且具有较好的查准率和查全率(特别对于含噪声图像).

Content-based image retrieval has become a significant research topic because of the proliferation of video and image data in digital form. Increased bandwidth availability to access the Internet in the near future will allow the users to search for and browse through video and image databases located at remote sites. Therefore fast retrieval of images from large databases is an important problem that needs to be addressed. The disadvantages of the traditional color image retrieval based on color histogram are not considering the color spatial distribution and high complexity of computation. And what＇s more, the retrieval results with the condition of noise image are not good as expected. So an efficient color image retrieval technique based on multi-features of bit-plane is proposed in this paper. Firstly, according to the noise attack characteristic, the significant bit-planes are extracted from the color image. Secondly, the weighted color histograms are extracted from the significant bit-planes as color feature, and the space information entropy of every significant bit-plane is computed as spatial feature. Finally, the similarity between color images is computed by using a combined index based on color feature and spatial feature. Experimental results show that the proposed color image retrieval is more accurate and efficient in retrieving the user-interested images. Especially, it can retrieve the noise （including fuzzy, sharpen, and illumination, etc） image effectively.