Analysis on Feature Extraction and Classification of Rice Kernels for Myanmar Rice Using Image Processing Techniques

The paper presents the analysis on feature extraction and classification of rice kernels for Myanmar rice using image processing techniques. There are seven steps to analyze the image processing techniques. The classification of export-rice quality is a great challenge in agricultural industries. The development of modern technology-based methods for rice quality classification is currently necessary to provide a reliable and consistent rice quality to consumers. An image processing algorithm has been implemented the analysis and classification the rice kernels in Myanmar. The proposed method is based on real-field feature and KNN classifier. Then, a series of measurements were done using image processing techniques on three classes of Paw-San rice in Myanmar. The real-field feature of Paw-San rice is percentage of broken rice contained in the batch. At 30 tests are conducted for each class of Paw-San rice. The simulation results show that the proposed method can confirm the classification accuracy in the range of 83~100% for the three grades.