基于颜色和边缘的快速图像检索研究

A Fast Image Retrieval Method based on Color and Improved Mathematical Morphology

在分析传统的图像检索技术基础上,研究了一种基于颜色和边缘特征的快速图像检索技术。首先,在HSI颜色空间中,使用色调直方图对原图像库进行初步检索,获得初级检索图像库;然后,用改进的数学形态学算法对初级检索图像库进行边缘提取,使用边缘像素点集合颜色直方图对初级检索图像库进行再次检索,从而得到最终的检索结果。实验结果表明方法在提高图像检索精度的同时,大大缩短了检索时间。

Based on detailed analysis oftraditional CBIR technology,thispaper presents a newmethod based on combination of the color and improved mathematical morphology for image retrieval.First of all,the candiate images are selected by using hue histogram among database images.Then,an improved mathematical morphology is used to extract the edge features in the candiate images obtained in the former step.Finally,thecolor histograms built up by edge features pixel dots are used to retrieve the candiateimages,thus obtaining the final retrieving results.Experimenal results show that this method is more efficient and less in time consumption.