欢迎访问《图学学报》 分享到:

图学学报

• 图像技术 • 上一篇    下一篇

基于内外边缘颜色特征的图像检索算法

  

  • 出版日期:2010-04-30 发布日期:2015-08-11

Image Retrieval Algorithm Based on the Color Feature of  Inner and Outer Edge

  • Online:2010-04-30 Published:2015-08-11

摘要: 算法的基本思路是提取边缘部分的颜色特征进行图像检索,将边缘分为内外边缘区别对待,加大内边缘(对应目标部分)的比重,通过多尺度多结构元的数学形态学方法实现。通过设置内外边缘的权重充分考虑了目标的重要性,克服了全局颜色直方图不能反映空间信息的缺陷。实验结果表明,该算法有效地提高了检索系统的效率。

关键词: 计算机应用, 图像检索, 数学形态学, 边缘颜色特征

Abstract: The basic idea of the algorithm is to extract the color feature of edge area for image retrieval. The edges are divided into inner edge and outer edge and the weight of inner edge (target part) is increased. This is realized through the method of mathematical morphology with multi-scale and multi-structuring elements. By setting a weight for inner and outer edge, the algorithm takes the important of the target into account and overcomes the shortcoming of color histogram which loses the location information.Experimental results show that this method improves the image retrieval performance effectively.

Key words: computer application, image retrieval, mathematical morphology;edge color feature