期刊名称：International Journal of Image Processing (IJIP)
电子版ISSN：1985-2304
出版年度：2013
卷号：7
期号：3
页码：278-285
出版社：Computer Science Journals
摘要：This paper will present an enhanced approach for the reconstruction of spectral reflectance by the combination between two methods, the Pseudo-Inverse (PI) as the base formula, whilst adaptively selecting the training samples as performed in the Adaptive Wiener estimation method proposed by Shen and Xin for the estimation of the spectral reflectance. This enhancement will be referred to as Adaptive Pseudo-Inverse (API) through this research. Training and verification datasets have been prepared from GretagMacbeth ColorChecker CC chart, Kodak Color Chart and a specially designed palette of Japanese organic and inorganic mineral pigments to test and compare the estimation results, using the Pseudo-Inverse and Adaptive Pseudo-Inverse method. The performance of spectral reconstruction methods will be presented in terms of spectral and colorimetric error for the estimation accuracy. The experimental results showed that the proposed method achieved better performance and noticeable decline in spectral estimation error.
