Non-equal Spacing Division of HSV Components for Wood Image Retrieval
To make the HSV color space more suitable for wood image retrieval, it analyzed hundred lumber images to educe their visual spatial distribution law in HSV color space, and in view of the fact that the hue, saturation and value component were all centralized distribution, it proposed that one kind of non-equal spacing division method on hue, saturation and value separately, according to human visual perception and histogram threshold segmentation. The result showed, the hue component could be divided into nine non-equal interval bins; the saturation component could be divided into four bins; and similarly, the value component could also be divided into four bins. Consequently, after carrying through the non-equal color space quantification of HSV space, it effectively compressed the dimension of feature vectors, meanwhile reduced the computational complexity of the features matching.
YU Hai-peng LUO Wei CAO Jun LIU Yi-xing
Northeast Forestry University, Harbin 150040, P.R.China
国际会议
International Symposium on Wood Science and Technology(IAWPS 2008)(国际木材产品学会联合会2008学术研讨会)
哈尔滨
英文
127-128
2008-09-27（万方平台首次上网日期，不代表论文的发表时间）