Electronic nose for odor classification of grains

Abstract

Cereal Chem. 73(4):457-461 An electronic nose was used to classify grain samples based on their networks. The samples were divided into either the four classes smell and to predict the degree of moldy/musty odor. A total of 235 sam- moldy/musty, acid/sour, burnt, or normal or the two classes good and bad ples of wheat, barley and oats, which had been odor classified by at least according to the inspectors descriptions. They were also assigned a score two grain inspectors, were used. Headspace samples from heated grain describing their intensity of moldy/musty odor. The electronic nose corwere pumped through chambers containing metal oxide semiconductor rectly classified =75% of the samples when using the four-class system field effect transistor (MOSFET) sensors, SnO 2 semiconductors and an and =90% when using the two-class system. These values exceeded the infrared detector monitoring CO2. The sensor signals were evaluated corresponding percentages of agreement between two grain inspectors with a pattern-recognition software program based on artificial neural classifying the grain. In Sweden, as well as in many other countries, grains are checked for off-odors upon delivery at granaries. Off-odors make grains and grain products less palatable and are often indicative of past or ongoing microbial deterioration. Thus off-odor characterization offers a potential way for quickly and cheaply assessing batches of grain to determine whether they should be accepted for human or animal consumption, used for other purposes, or rejected.