Image Retrieval Based on Multi-feature Fusion of Color Connected Region

XIANG Yang,CHONG Yan-wen,WANG Yi-nuo,PAN Shao-ming
DOI: https://doi.org/10.3969/j.issn.1000-3428.2012.15.059
2012-01-01
Abstract:Against the problem that the method of color coherence vector is easy to lose the spatial information of image content,a new method for image retrieval is proposed in this paper.It introduces the feature of narrow degree and standard deviation,proposes a new method for image block-dividing,the concept of centrifugal distance and a new distance formula.Specific practices are as follows.It extracts the feature of size,narrow degree and color of color connected regions.It extracts the number of pixels,the standard deviation and centrifugal distance of the image.It calculates the content similarity between images and retrieval.Experimental results in Corel image database show that this method can effectively use the spatial distribution information of the image,and accuracy is increased.
What problem does this paper attempt to address?