Major orthogonal dimensions measurement of food grains by machine vision using ImageJ

A machine vision ImageJ plugin was developed in Java for orthogonal length and width determination of singulated particles from digital images. A flatbed scanner obtained the digital images of particulate samples. The “pixel-march” method, which compared pixel colors to determine object boundaries for dimensional measurements, utilized only the ImageJ fitted-ellipse centroid coordinates and major axis inclination. The pixel-march started from objects centroid and proceeded along the fitted-ellipses' major and minor axes for boundary identification. Actual dimensions of selected reference particles measured using digital calipers validated the plugin. The plugin was applied to measure orthogonal dimensions of eight types of food grains. The plugin has overall accuracy greater than 96.6%, computation speed of 254 ± 125 particles/s, handles all shapes and particle orientations, makes repeatable measurements, and is economical. Applications of developed plugin may include routine laboratory dimensional measurements, physical dimensional characteristics, size based grading, and sieve analysis simulation for particle size distribution.