|Yao, Haibo - ITD|
|Hruska, Zuzana - ITD|
|Dicrispino, Kevin - ITD|
|Brabham, Kori - ITD|
|Lewis, David - ITD|
|Beach, Jim - ITD|
Submitted to: ASAE Annual International Meeting
Publication Type: Proceedings
Publication Acceptance Date: June 1, 2005
Publication Date: July 30, 2005
Citation: Yao, H., Hruska, Z., DiCrispino, K., Brabham, K., Lewis, D., Beach, J., Brown, R.L., Cleveland, T.E. 2005. Differentiation of fungi using hyperspectral imagery for food inspection. Proceedings of ASAE Annual International Meeting, July 17-20, 2005, Tampa, FL. Paper No. 053127. Interpretive Summary: Fungi grow almost everywhere and under most conditions. Some fungi are toxin-producing and pose a threat to humans and animals. Thus, it is important to detect and identify different toxin-producing molds. One benefit derived from fungal identification procedures is knowing which specific toxin is present in a given sample. Once a fungus is identified, it is possible to extrapolate the specific toxins that may be present as a consequence. For example, the presence of Aspergillus flavus does not necessarily indicate harmful levels of aflatoxin (toxin produced by Aspergillus spp.), however, it does mean that the potential for aflatoxin production is present. Another example is that some fungi, such as A. niger and Trichoderma viride,can inhibit the growth of A. flavus and subsequent aflatoxin detection. Thus, correct identification of a fungus can be helpful in isolating the presence of specific toxins associated with the identified organism. Furthermore, knowing which fungi exist in particular samples helps direct the future focus of detection technology. Applications of fungal detection/identification can be for food inspection, homeland security, household inspection, environmental protection, etc.
Technical Abstract: This paper is part of a project of using hyperspectral imagery to detect pathogens, such as mycotoxin-producing fungi, in grain products, such as corn. Traditionally, corn kernels have been examined for evidence of bright greenish-yellow fluorescence (BGYF), indicative of the presence of A. flavus, when illuminated with a high-intensity ultra-violet light. The BGYF approach is time- and labor-intensive, and somewhat inaccurate. Several previous studies have examined spectral-based, non-destructive methods for the detection of fungi and toxins. This research focuses on using spectral image data for fungi and toxin detection. A tabletop hyperspectral imaging system, VNIR-100E, is used in the study for high spectral and high spatial resolution spectral data acquisition. In this paper, a total of five toxin producing fungal species were used in two experiments. They are Penicillium chrysogenum, Fusarium moniliforme, Aspergillus parasiticus, Trichoderma viride, and Aspergillus flavus. All fungal isolates were cultured on agar in Petri dishes under lab conditions and were imaged on day 5 of growth. The objective of this study is to use hyperspectral imagery for classification of different fungi.