Abstract：A colorimetric sensor array was developed using metalloporphyrins. Odors of peaches of three different classes （normal, physical damage, fungal diseases） were detected. The RGB value changes of the sensor array before and after exposure to the headspace gas of peaches were analyzed by principal component analysis （PCA）. The former 6 principal components were selected for the model building of the least squares support vector machine （LS -SVM）. The discriminating rate of the model is 100% for 26 samples in training set, and 91.7% for 24 samples in predicting set. Experiments show that it is feasible to apply the colorimetric sensor array to evaluate the quality of the peaches. The method developed can also be used as reference for quality evaluation of other fruits.
2009, Vol. 30 
