一种用于彩色目标检测的混合颜色模型

A Mixed Color Model to Detect Color Target

针对光照变化条件下彩色目标的鲁棒检测要求,分析了单独采用HSV和L*a*b*颜色空间检测彩色目标的不足之处,为进一步降低光照变化和背景干扰,提出采用改进的紧凑彩色模型。考虑到不同颜色在HSV和L*a*b*颜色空间中的分布差异,建立了一种联合这2种颜色空间紧凑彩色分布特征的混合颜色模型,用于更准确地描述目标的颜色特征。试验结果表明,使用混合颜色模型可以获得更好的目标检测效果。

By the requirement of color target detection in variable illumination environment, the respective shortage of the HSV color space and L＊a＊b＊ color space was analyzed. To reduce the effects from variable lighting conditions and background color distribution, two compact color models were generated. Furthermore, considering the color distribution difference in the both color spaces, a mixed color model based on the two compact color models was proposed to distinguish target color from similar background color in this paper. Experimental results showed that a better detection result can be obtained when the mixed color model is used.