【作者】 张哲超；

【导师】 沈会良；

【作者基本信息】 浙江大学 ， 电子科学与技术， 2010， 硕士

【摘要】 随着计算机图像处理技术的飞速发展,彩色图像和多光谱图像越来越多地应用于颜色信息的展现、传递以及复制。不同的成像设备具有各自的色彩特征,而且所获得及传递的颜色信息都是设备相关的。为了能够精确地得到颜色信息,需要对成像设备进行色彩表征,得到与设备无关的所成像物体的光谱反射率。本文对在非线性成像系统下的光谱反射率重建方法和在反射率重建过程中代表颜色的选取方法进行了研究。当成像系统为非线性或测量过程受到噪声干扰时,已有的光谱反射率重建方法都无法得到较高的重建精度。例如,线性的维纳估计法无法很好地处理非线性的问题。为了很好地处理系统非线性和测量噪声,本文研究了文献中结合多项式的普通最小二乘法和正则化最小二乘法。实验结果表明,在光谱误差和色度误差方面,正则化最小二乘法明显优于维纳估计法和普通最小二乘法。色彩表征过程中通常会用到标准色卡,但由于标准色卡的颜色样本数量较大,在实际成像系统中使用时有诸多不便。考虑到颜色样本之间存在较大冗余,可从中选出具代表性的少数样本用于光谱表征。本论文提出了一种代表颜色的分步选取算法,即首先通过假设一个虚拟成像系统,根据全局误差最小的原则,挑选出部分最具代表性的颜色,估计出实际成像系统的光谱响应函数,然后在此基础上继续选择其余的代表颜色。实验表明,本论文所提出的方法在光谱精度及色度方面均优于先前方法。更多还原

【Abstract】 With the rapid development of modern computer image processing technologies, color images and multispectral images have been widely used in the presentation, transfer and copying of color information. Different imaging systems have different characteristics, and all the color information they get and transfer are device-dependent, so it is necessary to find a way to get the device-independent color information. Spectral reflectance reconstruction, also referred as spectral characterization, aims to recover accurate spectral reflectance of object surface by employing standard color charts. This dissertation studies the methods for spectral reflectance reconstruction and the methods for selection of representative colors.When the imaging process is not a linear system or influenced by the measurement noise, the spectral reconstruction methods in hand cannot work very well. For instance, Wiener estimation only works well under linear systems. In order to cope with the nonlinearity and measurement noise, we studied the polynomial regression solved by ordinary least squares and regularized least squares. Experiment results show that, in terms of spectral and colorimetric error metrics, the regularized method performs better than Wiener estimation and ordinary polynomial regression.As there are always a large number of color samples on a color chart, spectral characterization becomes a time-consuming process for practical application. Some methods have been presented to selected representative color samples based on the redundancy of the colors on a chart. However, these methods only consider the distribution of spectral reflectance, and thus the selected colors may not be optimal for a specific imaging system. To deal with this problem, we proposed a sequential method for the selection of most representative colors, which consists of two steps. In the first step, a part of representative colors is selected according to the minimization of mean spectral root-mean-square error, by assuming a virtual imaging system. The spectral sensitivity of the real imaging system is then calculated based on these selected samples. In the second step, additional representative colors are selected based on the characteristics of the real imaging system. Experiment shows that the proposed method outperforms the previous methods in terms of both spectral and colorimetric accuracy.更多还原