Abstract
Brown blotch, caused by pathogenic Pseudomonas tolaasii, is the most problematic bacterial disease in Agaricus bisporus mushrooms. Although it does not cause any health problems, it reduces the consumer appeal of mushrooms in the market place, generating important economic losses worldwide. Hyperspectral imaging (HS) is a non-destructive technique that combines imaging and spectroscopy to obtain information from a sample. The objective of this study was to investigate the use of HSI for brown blotch identification and discrimination from mechanical damage on mushrooms. Hyperspectral images of mushrooms subjected to (1) no treatment, (2) mechanical damage or (3) microbiological spoilage were taken during storage and spectra representing each of the classes were selected. Partial least squares-discriminant analysis (PLS-DA) was carried out in two steps: (1) discrimination between undamaged and damaged mushrooms and (2) discrimination between damage sources (i.e. mechanical or microbiological). The models were applied at a pixel level and a decision tree was used to classify mushrooms into one of the aforementioned classes. A correct classification of >95% was achieved. Results from this study could be used for the development of a sensor to detect and classify mushroom damage of mechanical and microbial origin, which would enable the industry to make rapid and automated decisions to discard produce of poor marketability.
© 2010 IM Publications LLP
PDF Article
More Like This
Cited By
You do not have subscription access to this journal. Cited by links are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.
Contact your librarian or system administrator
or
Login to access Optica Member Subscription