ABSTRACT
Aflatoxin is a kind of highly toxic and carcinogenic substance with the characteristic of ultraviolet (UV) fluorescence. This chapter discusses the detection of artificial aflatoxin contamination using a hyperspectral imaging system based on the aflatoxin’s fluorescence characteristic under UV light. An online fast aflatoxin detection method is found through the optimization of fluorescence index and narrowband spectra. Support vector machine was first proposed by Corinna Cortes and Vapnik in 1995. It has shown many special advantages in solving small-sample cases and nonlinear high-dimensional pattern recognition problems. A correlational study shows that gray-level quantitative value of aflatoxin may provide meaningful feature for aflatoxin detection and the gray histogram distribution changes with different aflatoxin contents. The recognition effect of classifier is heavily dependent on the quality of feature extraction. The purpose of feature extraction is to reduce the dimensionality curse and the feature dimension to a reasonable range, which can improve the performance of the classifier and increase the recognition speed.
