Feature reduction of Zernike moments using genetic algorithm for neural network classification of rice grain
Chong Yaw Wee, P. Raveendran, Fumiaki Takeda, T. Tsuzuki, H. Kadota, S. Shimanouchi · 2003
In this paper, Zernike moment features extracted from rice grains are used in classifying normal and damaged rice. Genetic algorithm (GA) is used to reduce the number of features while maximizing the classification performance. The GA chromosome fitness is evaluated using a multilayer perceptron (MLP) trained by backpropagation learning algorithm.