Grading agricultural products with machine vision

Abstract

Three applications of using machine vision to grade agricultural products are presented: grading container-grown ornamental plants; predicting the time of molting for soft-shelled crawfish; and detecting cracks in milled and brown rice. The soft-shelled crawfish system was able to predict the time of molting within three days for 75 to 85 percent of the crawfish. The ornamental plant grader was not able to match the human grader standards, but is was able to provide an objective set of measurements describing several plant features. The system to measure cracks had an accuracy of 86.5 percent for brown rice and 92.3 percent for milled rice. All three systems have the potential to be used with robotics and/or automation for grading and/or sorting operations.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>