首页>
外国专利>
Color coordinate space structure and color quantization method using this color coordinate
Color coordinate space structure and color quantization method using this color coordinate
展开▼
机译:颜色坐标空间结构和使用该颜色坐标的颜色量化方法
展开▼
页面导航
摘要
著录项
相似文献
摘要
The present invention relates to a color coordinate space structure for content-based image retrieval and a color quantization method using the color coordinates.;In the conventional HSV color space, a change in color that a human's vision feels in the same space appears differently. In this case, if the color is simply quantized in the HSV color space, the quantized colors do not express all colors evenly. The difficulty of vector quantization and the increase of computation is one factor that degrades the content-based image retrieval performance.;As shown in FIG. 1, the present invention proposes a new color space structure composed of color (Hue = h o ), maximum value (Max = max), minimum value (Min = min), and difference value (Difference = d). And a color coordinate space structure for performing color quantization (mapping M colors into N colors; M N, M and N are positive integer values) using coordinate values in the color space, and color quantization using the color coordinates. By providing a method, it is possible to consider the visual characteristics of a person and to perform high speed digital processing of color mapping.
展开▼
机译:本发明涉及一种用于基于内容的图像检索的色坐标空间结构和使用该色坐标的颜色量化方法。在传统的HSV色空间中,人的视觉在相同空间中感觉到的颜色变化以不同的方式出现。在这种情况下,如果仅在HSV颜色空间中对颜色进行量化,则量化后的颜色不能均匀地表示所有颜色。矢量量化的困难和计算的增加是降低基于内容的图像检索性能的一个因素。参照图1,本发明提出了一种新的颜色空间结构,该结构由颜色(Hue = h o Sup>),最大值(Max = max),最小值(Min = min)和差值(Difference = d)。并且,颜色坐标空间结构用于使用颜色空间中的坐标值进行颜色量化(将M种颜色映射为N种颜色; M N,M和N为正整数值),并使用颜色坐标进行颜色量化。通过提供一种方法,可以考虑人的视觉特征并执行颜色映射的高速数字处理。
展开▼