Journal article
Effect of Spectral Resolution on Pattern Recognition Analysis Using Passive Fourier Transform Infrared Sensor Data
Applied spectroscopy, Vol.53(11), pp.1382-1391
11/1999
DOI: 10.1366/0003702991945722
Abstract
The Fourier transform infrared (FT-IR) spectral data of two nerve agent simulants, diisopropyl methyl phosphonate (DIMP) and dimethyl methyl phosphonate (DMMP), are used as test cases to determine the spectral resolution that gives optimal pattern recognition performance. DIMP is used as the target analyte for detection, while DMMP is used to test the ability of the automated pattern recognition methodology to detect the analyte selectively. Interferogram data are collected by using a Midac passive FT-IR instrument. The methodology is based on the application of pattern recognition techniques to short segments of single-beam spectra obtained by Fourier processing the collected interferogram data. The work described in this article evaluates the effect of varying spectral resolution on the pattern recognition results. The objective is to determine the optimal spectral resolution to be used for data collection. The results of this study indicate that the data with a nominal spectral resolution of 16 cm−1 provide sufficient selectivity to give pattern recognition results comparable to that obtained by using higher resolution data. We found that, while higher resolution does not increase selectivity sufficiently to provide better pattern recognition results, lower resolution decreases selectivity and degrades the pattern recognition results. These results can be used as guidelines to maximize detection sensitivity, to minimize the time needed for data collection, and to reduce data storage requirements.
Details
- Title: Subtitle
- Effect of Spectral Resolution on Pattern Recognition Analysis Using Passive Fourier Transform Infrared Sensor Data
- Creators
- Arjun S Bangalore - Chemical Technology Division, (A. S. Bangalore, A. S. Boparai) and Environmental Research Division (J.C.D.), Argonne National Laboratory, 9700 S. Cass Avenue, Argonne, Illinois 60439; and Center for Intelligent Chemical Instrumentation, Department of Chemistry, Clippinger Laboratories, Ohio University, Athens, Ohio 45701-2979 (G.W.S.)Jack C Demirgian - Chemical Technology Division, (A. S. Bangalore, A. S. Boparai) and Environmental Research Division (J.C.D.), Argonne National Laboratory, 9700 S. Cass Avenue, Argonne, Illinois 60439; and Center for Intelligent Chemical Instrumentation, Department of Chemistry, Clippinger Laboratories, Ohio University, Athens, Ohio 45701-2979 (G.W.S.)Amrit S Boparai - Chemical Technology Division, (A. S. Bangalore, A. S. Boparai) and Environmental Research Division (J.C.D.), Argonne National Laboratory, 9700 S. Cass Avenue, Argonne, Illinois 60439; and Center for Intelligent Chemical Instrumentation, Department of Chemistry, Clippinger Laboratories, Ohio University, Athens, Ohio 45701-2979 (G.W.S.)Gary W Small - Chemical Technology Division, (A. S. Bangalore, A. S. Boparai) and Environmental Research Division (J.C.D.), Argonne National Laboratory, 9700 S. Cass Avenue, Argonne, Illinois 60439; and Center for Intelligent Chemical Instrumentation, Department of Chemistry, Clippinger Laboratories, Ohio University, Athens, Ohio 45701-2979 (G.W.S.)
- Resource Type
- Journal article
- Publication Details
- Applied spectroscopy, Vol.53(11), pp.1382-1391
- DOI
- 10.1366/0003702991945722
- ISSN
- 0003-7028
- eISSN
- 1943-3530
- Language
- English
- Date published
- 11/1999
- Academic Unit
- Chemistry
- Record Identifier
- 9984216601102771
Metrics
2 Record Views