Effects of Storage Duration on Detecting Watercore in Apples Using Machine Vision

Abstract

Measuring the amount of light transmitted through an apple has had limited success for detecting watercore in apples when the measurements were made soon after harvest. However, errors associated with this technique when applied to apples coming out of cold storage has yet to be determined. Using a machine vision system to view the stem-end of the apple while illuminating the calyx, the intensity of light transmitted through individual apples was measured biweekly for 14 weeks in 1990. The mean gray-level of the pixels within a square window centered about the stem was representative of the intensity of the light transmitted through the fruit. The intensity of light transmitted through the fruit decreased with time for all classes of watercored fruit. Based upon the initial levels defined for each class, this decrease in light transmission resulted in misclassifying watercored fruit as watercore-free at the end of the time study. Of 72 fruit initially classified as severely watercored, 18% of the fruit were misclassified as watercore-free, because three apples still had evidence of watercore, while eight fruit exhibited internal browning, due to watercore.