Edge Detection Techniques for Rice Grain Quality Analysis using Image Processing Techniques

Abstract

In agricultural countries like the Philippines, rice grain is considered the most important crop in the world for human consumption as daily food and in the food market, thus quality control must be considered. Rice grain quality evaluation is done manually, which is non-reliable, time-consuming and costly. The quality of rice grain is categorized by the combination of physical and chemical characteristics. Grain appearance, color, size and shape, chalkiness, whiteness, degree of milling, bulk density, foreign matter content, and moisture content are some physical characteristics, while amylose content of the endosperm, gelatinization temperature of the endosperm starch, and Na content are chemical characteristics. This paper presents a solution for the grading and evaluation of rice grains on the basis of grain size and shape using Scilab Image Video Progressing (SIVP) techniques. Specifically, an edge detection algorithm is used to find out the region of the boundaries of each grain. This method requires a minimum of time and is more affordable. Edge detection is vital for its reliability and security, as well as for providing a better understanding of automatic identification in computer vision applications. This study determines the best techniques among the edge detection algorithms.