Fusarium head blight is a fungal disease that affects the world's small grains, such as wheat and barley. Attacking the spikelets during development, the fungus causes a reduction of yield and grain of poorer processing quality. It also is a health concern because of the secondary metabolite, deoxynivalenol, which often accompanies the fungus. While chemical methods exist to measure the concentration of the mycotoxin and manual visual inspection is used to ascertain the level of Fusarium damage, research has been active in developing fast, optically based techniques that can assess this form of damage. In the current study a near-infrared (1000-1700 nm) hyperspectral image system was assembled and applied to Fusarium-damaged kernel recognition. With anticipation of an eventual multispectral imaging system design, 5 wavelengths were manually selected from a pool of 146 images as the most promising, such that when combined in pairs or triplets, Fusarium damage could be identified. We present the results of two pairs of wavelengths [(1199, 1474 nm) and (1315, 1474 nm)] whose reflectance values produced adequate separation of kernels of healthy appearance (i.e., asymptomatic condition) from kernels possessing Fusarium damage.