Research of Rice Appearance Quality Recognition

Rice is the common food in people's daily life, different grades of rice have a great difference to the influence of taste, therefore, it is important to identify the rice quality. In order to improve the accuracy and real-time performance of the rice quality identification, PCA algorithm (Principal Component Analysis) and BP neural network algorithm are proposed to identify the rice quality. The feature variables of rice are extracted by PCA algorithm, then training the BP neural network parameters with the variables as its input, finally the BP neural network trained is used to identify the rice quality. The algorithm is implemented into MATLAB software. The simulation shows that the PCA-BP algorithm is greatly well than BP algorithm in accuracy and real-time performance.