Potato Grading Based on Size Features by Machine Vision Technique

Abstract

In this study, a method was established to grade the potato based on size features using a machine vision technique with image processing and multivariate analysis method. The image of individual potato was captured using a color camera sensor with white LED lighting conditions in the laboratory. An image processing algorithm was developed for extracting the size (major, minor, and surface area) features from 57 potato images. Using these extracted feature data, the potatoes were classified using the partial least squares-discriminant analysis (PLS-DA) algorithm, and the overall accuracy was achieved at 86%. Finally, it was stated that the PLS-DA classification algorithm with size features could be used for grading the potato in Bangladesh.