Development of Apple MRI Dataset for Internal Quality Analysis

Abstract

Internal quality assessment of agricultural products is a challenging task for exporting premium quality agri products like apple fruits. In this paper, we have analyzed the internal quality of apple fruits by a non-destructive method. We have developed our dataset of MR images of apples by subjecting 21 apples to MRI scanning for the development of robotics detection of internal defects in apples. This MRI scanning led to 196 MR images. A comparative study was carried out based on MRI images with respective external photographic images of apple fruits. Depending on external as well as internal defects the apples were grouped into four categories. Through this study, we can easily identify the percentage and area of the defect without affecting the physical appearance of the apple.