基于颜色一致向量的图像检索方法

大量图像信息的产生使得基于内容的图像检索技术成为研究热点。由于颜色特征具有稳定性和计算简单的特点，本文首先介绍了利用全局颜色直方图进行图像检索的基本思想，然后分析了它的局限性，并给出了改进的方法：特征提取采用结合空间信息的颜色一致向量方法。在特征度量时，依据所设计的评价实验，对这两种方法进行了比较，并给出了实验结果和图像检索性能的评价。实验表明，所述的图像检索方法具有较好的查全率和查准率。

An Approach to Image Retrieval Based on Color Coherence Vector

The creation of large image information makes the content-based image retrieval become one of the most active research areas. Based on the stability of the color feature and its simpleness of computing, this paper introduces the technique of image retrieval based on global color first, also discusses and analyzes some deficiencies about the method, and presents some improved approaches that a new image retrieval feature called the color coherence vector is adopted, which incorporates the spatial information and uses it for image indexing and comparison. As for distance measurements, an experiment is designed to evaluate the effectiveness of the two kinds of methods, and experimental results are reported and presented to demonstrate the effectiveness and efficiency of the proposed approach. It has been proved that the method can get better recall ratio and precision ratio.