Spectrum is the essence of color as it contains the most precise information about a color’s components. The liquid crystal display (LCD) spectrum is ubiquitously used in the color science field, so there is a great demand for fast and high-precision spectral characterization of color. Simulated Multispectral Imaging Reconstruction Technique with Weighted Principal Component Analysis (SMIRTWPCA) for fast characterization of LCD is proposed. The spectral data are compressed by color weighted principal component analysis (PCA), and the three-component image (RGB) channel is expanded by a polynomial to simulate the multi-channel characteristics of the multi-spectral image. The expanded RGB channel and compressed spectral data are used as input and output variables for Wiener estimation to calculate the transfer coefficient matrix. Any RGB can obtain its low-dimensional spectral data through the optimal expansion and transfer coefficient matrix and then reconstruct the spectrum through weighted PCA to realize RGB to spectral characterization. The RMSE and CIEDE2000 color differences were used to evaluate the accuracy of this method. Compared with other spectral characterization models for LCD, the experimental results show that the average RMSE of the SMIRTWPCA method is 3.76E-06, and the average of the CIEDE2000 color difference method is 0.61ΔE00, which indicate that the method is suitable for LCD spectral characterization.