Abstrakt

Medical imaging based spectral reflectance reconstruction combining PCA and regularized polynomial

Ke Wang, Huiqin Wang, Zhan Wang, Ying Yin, Yi Zhang

A spectral reflectance reconstruction method based on combining PCA and regularized polynomial is proposed to optimize the spectral data and channel response. Firstly, the PCA method is used to reduce the high dimensional spectral data of the training samples. Then, the polynomial expansion for channel response of the sample is carried out to improve the accuracy of spectral reconstruction through the PCA reduction dimension. At the same time, Tikhonov restriction is added to avoid data instability and random noise caused by polynomial expansion. The simulation experimental results show that, the method presented in this paper is better than the previous method in three precision evaluation method (RMSE, GFC, ISSD).