基于有限维线性模型的监督颜色回归恒常性算法

A Supervise Color Constancy Algorithm Based on Finite-dimension Linear Model

提出了一种监督颜色回归恒常性算法 ,可将某种光照条件下获得的颜色值转换到事先确定的标准光照下的颜色值 ,该算法基于表面反射的有限维线性模型 ,利用多元线性回归方程的最小二乘法求解转换矩阵和修正矢量。给出了算法的基本思想和实验结果 ,同以前的算法相比 ,本算法应用条件几乎不受限制 。

Eliminating the influence of illumination is very important in applications such as color image recognition and analysis. In this paper, we proposed a new supervise color constancy algorithm based on finite dimension linear model. It can transform the color values under the abnormal illuminations into a normal illumination. Comparing with the previous methods, the algorithm does not need to know the spectral reflectance functions of color chips and the camera spectral sensitivity, thus can be efficiently applied in practice. We give the main design of the algorithm, its implementation in real imaging systems, and the experimental results.