Author
YAO, HAIBO - Mississippi State University | |
HRUSKA, ZUZANA - Mississippi State University | |
KINCAID, RUSSELL - Mississippi State University | |
ONONYE, AMBROSE - Mississippi State University | |
Brown, Robert | |
Cleveland, Thomas |
Submitted to: Workshop Proceedings
Publication Type: Proceedings Publication Acceptance Date: 3/11/2010 Publication Date: 7/12/2010 Citation: Yao, H., Hruska, Z., Kincaid, R., Ononye, A., Brown, R.L., Cleveland, T.E. 2010. Spectral Angle Mapper Classification of Fluorescence Hyperspectral Image for Aflatoxin Contaminated Corn. Proceedings of IEEE 2nd Workshop on Hyperspectral Image and Signal Processing: Evolution in Remote Sensing Conference. p. 129. Interpretive Summary: Technical Abstract: Aflatoxin contamination in corn is a serious problem for both producers and consumers. The present study applied the Spectral Angle Mapper classification technique to classify single corn kernels into contaminated and healthy groups. Fluorescence hyperspectral images were used in the classification. Actual corn aflatoxin contamination was chemically analyzed using the VICAM analytical method. An obvious fluorescence peak shift was observed to be associated with the aflatoxin contaminated corn. Aflatoxin classification levels were based on FDA’s regulation, including 20 ppb (parts per billion) for human consumption and 100 ppb for feed. Classification accuracy for the 20 ppb level is 86% with a false positive rate of 15%. For the 100 ppb level, the accuracy is 88% with a false positive rate of 16%. The results indicate that the Spectral Angle Mapper method and fluorescence hyperspectral imagery have the potential to classify aflatoxin contaminated corn kernels. |