Classification of rice kernels using wavelet packet transform and support vector machine

A classification algorithm was developed to differentiate individual infected (dead, chalky, cracked, and immature) and qualified rice kernels. The image was preprocessed by wavelet packet, and the feature regions of interest were extracted by edge detection. Ten statistical features (area, perimeter, compactness, etc.) were extracted from the image data of single kernels. The statistical features composed the pattern vector of a single kernel. The dimensionality of pattern vectors was reduced by principal component analysis. A multi-class support vector machine with kernel of radial basis function was used for classification. Using the statistical features, the rice kernels infected by dead, chalky, cracked, and immature and healthy rice kernels were classified with accuracies of 95.7%, 91.6%, 99.8%, 96.8% and 100%, respectively. Almost perfect classification was obtained under the infected vs. healthy model.