The Determination of Morphological and Textural Features of Rice Grains in a sample using Digital Image Processing Technology and the Classification of the Rice Grains in the sample

摘要

The quality of rice, the world's most important staple food crop, can be determined based on the shape, size, and texture of the grain. This paper discusses the methodology used in the development of a software in Visual C++ which can extract the physical features of rice grains in a sample using Digital Image Processing Technology. Image acquisition is done by spreading a handful of grains obtained from the sample on a dark background and obtaining its digital image using either a scanner or a digital camera. This image (in the BMP file format) is then used to extract the morphological and textural features of the grains using various image processing techniques. Each grain is described in terms of parameters such as length, breadth, area, perimeter, texture and colour. An analysis of these morphological and textural parameters allows for the classification of the rice grains into healthy and defective grains, thus determining the quality of the sample. Our method also includes the determination of the degree of polish of a sample of rice grains.