QUALITY CLASSIFICATION OF GRAIN USING A SENSOR ARRAY AND PATTERN-RECOGNITION

Measurements using arrays of electrochemical gas sensors, combined with pattern recognition methods, were used to classify wheat samples by quality grade. The classifications corresponded closely to those made by trained grain inspectors. Volatile compounds evolved from warmed samples of grain were passed over a heated noble metal catalyst and then into a series of electrochemical sensors. Signals from four sensors were recorded for four different catalyst temperatures in order to generate 16 signals for each grain odor sample. The 16 sensor signals were treated as a 16-dimensional vector or pattern of responses that was characteristic of the odor sample. The patterns for different grain odor samples were compared using both nearest-neighbor analysis and a commercial neural network simulation (NNS) program. These methods classified the samples correctly by grade with an accuracy of 68% and 65%, respectively. After compensation for instrument parameters, the NNS score improved to 83%; the nearest-neighbor analysis could not be similarly compensated. The robustness of the two algorithms was compared by adding simulated random and systematic errors to the sensor response patterns. The original data were used as the training set, and the patterns with errors added were used as the test set. In these cases, the NNS consistently outperformed the nearest-neighbor method at classification of the grain odor samples.