Prediction of Tomato Freshness Using Infrared Thermal Imaging and Transient Step Heating
Conference Paper
Overview
Research
Identity
Additional Document Info
Other
View All
Overview
abstract
2016 SPIE. Tomatoes are the world's 8th most valuable agricultural product, valued at $58 billion dollars annually. Nondestructive testing and inspection of tomatoes is challenging and multi-faceted. Optical imaging is used for quality grading and ripeness. Spectral and hyperspectral imaging are used to detect surface detects and cuticle cracks. Infrared thermography has been used to distinguish between different stages of maturity. However, determining the freshness of tomatoes is still an open problem. For this research, infrared thermography was used for freshness prediction. Infrared images were captured at a rate of 1 frame per second during heating (0 to 40 seconds) and cooling (0 to 160 seconds). The absolute temperatures of the acquired images were plotted. Regions with higher temperature differences between fresh and less fresh (rotten within three days) tomatoes of approximately uniform size and shape were used as the input nodes in a three-layer artificial neural network (ANN) model. Two-thirds of the data were used for training and one-third was used for testing. Results suggest that by using infrared imaging data as input to an ANN model, tomato freshness can be predicted with 90% accuracy. T-tests and F-tests were conducted based on absolute temperature over time. The results suggest that there is a mean temperature difference between fresh and less fresh tomatoes ( = 0.05). However, there is no statistical difference in terms of temperature variation, which suggests a water concentration difference.
