A Novel Hybrid Approach of KPCA and SVM for Crop Quality Classification

Abstract

Quality evaluation and classification is very important for crop market price determination. A lot of methods have been applied in the field of quality classification including principal component analysis (PCA) and artificial neural network (ANN) etc. The use of ANN has been shown to be a cost-effective technique. But their training is featured with some drawbacks such as small sample effect, black box effect and prone to overfitting. This paper proposes a novel hybrid approach of kernel principal component analysis (KPCA) with support vector machine (SVM) for developing the accuracy of quality classification. The tobacco quality data is evaluated in the experiment. Traditional PCA-SVM, SVM and ANN are investigated as comparison basis. The experimental results show that the proposed approach can achieve better performance in crop quality classification.