Temperature And Emissivity Separation From Tasi Data Based On Wavebands Selection

The TASI instrument has 32 spectral band covering 8 mu J m similar to 11.5 mu m. The main goal of this paper is to analyze the SNR of each band and the temperature accuracy based on wavebands selection. SNR is calculated by RLSD algorithm. The TES algorithm selects three empirical relationships (MMD, MMR, and VAR). The band of least SNR was removed one by one from residual bands, and the temperature errors were computed. The results show that temperature accuracy of TES using 17-20 wavebands is highest for MMR and VAR algorithm, but it is different from above case. So it is important to evaluate the quality of thermal images, and remove the high-noises wavebands for MMR and VAR relationship before TES.