Fresh fruit is perishable and difficult to preserve because of its high moisture content. Drying is one of the oldest, most effective, and most used processing technologies, which could reduce the moisture content to extend the shelf life by inhibiting microbial growth and reducing the rate of deteriorative reactions. Drying is an energy-consuming process, especially for hot air drying, which has the advantage of low cost and simple equipment compared to other drying methods, but the drying time is longer.
To maintain the product quality is still a challenge for most drying industries. Currently, the inspection of the appearance of dehydrated fruit products mainly relies on the trained human eye. This method is time-consuming and highly subjective, resulting in inaccurate results. On-line Computer vision systems have considerable potential to replace the human eye in industrial production. It is not only suitable for classifying the appearance of the final product but also makes it possible to predict and control the drying process.
In this project, we will try our first step about on-line computer visions systems. Mathematical models connecting the fruit images to their quality characteristics such as appearance and shrinkage will be built using machine learning methods such as Artificial Neural Network (ANN).
Undergraduate
(1) To investigate the connection between fruit drying characteristics (i.e. appearance) and images.
(2) To build mathematical models predicting the fruit drying charatersistics.
None
No lab has been assigned to this project