Computer vision for real-time monitoring of shrinkage for peas dried in a fluidized bed dryer

Abstract

This research work was focused on the use of computer vision (CV) to monitor in real-time, the shrinkage of peas drying in a fluidized bed dryer. The experiment was conducted at a constant air velocity of 10 m/s, and at temperature levels of 50, 55, and 60 °C, respectively. Shrinkage was monitored separately with a real-time and an offline CV system. The monitoring was carried out by measuring the area, perimeter, and the diameter of peas at specific intervals during the drying process. A real-time CV algorithm has been developed to compute the average area, perimeter, and diameter at each interval, by detecting and analyzing only seeds that were within the region of focus of the camera. The result from the offline method showed a steady increase in shrinkage as drying progressed. Similarly, the results from the real-time shrinkage curve, from 0 to 15 min were similar to those on the offline curve. At the end of the study, it was noted that areas of improvement which could enhance the results of the real-time method are needed.