Advanced Signal Processing and Pattern Recognition Methods for Passive Infrared Remote Sensors

摘要

This report describes studies directed to the automated analysis of Fourier transform infrared remote sensing interferogram data. The research presented here demonstrates that both qualitative and quantitative information can be extracted from short segments of digitally filtered interferograms without the need for any background or reference measurement. Through the use of experimental design techniques, an optimization protocol is devised for determining the key implementation parameters of the interferogram based analysis. An automated compound identification algorithm is developed for trichloroethylene (TCE) and is shown to operate effectively in the presence of a wide variety of infrared backgrounds. A quantitative analysis for sulfur dioxide (SO2) is also implemented through the direct use of short interferogram segments.