Computer Vision System for Ginseng Drying

A control system for multi-stage ginseng drying, based on real-time identification of moisture and quality changes, was developed. Computer-vision system provided real-time imaging of area shrinkage and colour changes. Shrinkage was identified by extracting of the green plane from the RGB color space, while colour changes- from the HSI color space. Shrinkage-moisture and colour-quality relationships were experimentally developed over the entire range of drying conditions in isothermal and non-isothermal drying. Relational models were used as components of control system. Data from computer vision system were used to generate relevant signals to low-level controller for temperature adjustment according to actual moisture content and quality. It enabled to start drying with the predictive model and then re-evaluate the drying conditions on the basis of feedback information from computer-vision system. Testing of a pilot batch dryer with an embedded virtual controller showed robustness and reliability of control, which provided drying to target moisture content 0.1 g/g (db) with high-standard quality.

Alex Martynenko is professor in Bioelectronics, Department of Engineering, Nova Scotia Agricultural College. Working on the border of engineering and biological science, he is developing innovative instrumentation for agriculture and environmental applications, including computer-vision for value-added processing of biomaterials.