Rice grain classification using image processing technique

Abstract

Rice is one of the most widely consumed staple foods, especially in the Asian subcontinent. The quality and varieties of rice grains identification is necessary to avoid mislabeling of rice grain varieties. Mostly it has done by visually. In manual classifications features like Major Axis, Minor Axis, Perimeter, Area, Aspect Ratio, Eccentricity, and Shape Factor are measured by using the specialized tools like calipers and other specialized tools. These features are feed into the machine learning techniques. It is a time consuming process and also there is a chance for measurement errors. In this proposed work the rice images are captured by camera and some preprocessing is done on the image to enhance the image quality. Feature extraction is performed on the collected image using image processing method through MATLAB. The extracted features are given to the Support Vector Machine for classification. The proposed work provides an improved classification accuracy up to 96% with minimum process time.