Aspect Ratio Analysis Using Image Processing for Rice Grain Quality

Abstract

Determination of aspect ratio distribution is important for elongated, needle-shaped particles whose utility and/or value may depend on this feature. In this work rice grain is taken as an example of such a particle and its aspect ratio distribution in various samples is found using image processing. The samples examined were from three different grades (commonly termed as full, half and broken) sold in local market and priced according to their size. From the analysis, reference aspect ratios were assigned to classify the grains and hence determine the extent of off-size in each market grade. Further, the effectiveness of the technique to quantify mixed or adulterated grades was studied. It was found that it is possible to know the undesired content within 10 percent accuracy.