Detecting Fusarium head blight in wheat kernels using hyperspectral imaging

Because of the health risks associated with the ingestion of the mycotoxin deoxynivalenol (DON) produced by Fusarium head blight (FHB), improving its detection in wheat kernels is a major research goal. Currently, assessments are largely performed visually by human experts. Being subjective, such assessments may not always be consistent or entirely reliable. As a result, methods with a higher degree of objectivity have been investigated, and special attention has been dedicated to the use of hyperspectral imaging (HSI) as the basis for more reliable detection strategies. This paper presents an algorithm for automatic detection of FHB in wheat kernels using HSI. The goal was to develop a simple and accurate algorithm which gave as output an index that can be interpreted as the likelihood of the kernel being infected by FHB. With a classification accuracy above 91%, the developed algorithm was robust to factors such as shape, orientation, shadowing and clustering of kernels. It was shown that the algorithm was not only suitable for detecting FHB, but it also has the capability, albeit limited, of estimating DON concentrations in wheat kernels.