Hyperspectral Imaging Techniques for Quality Assessment in Fresh Horticultural Produce and Prospects for Measurement of Mechanical Damage

Traditional techniques for quality assessment of fresh agricultural products are onerous and costly. Imaging systems based on the visible wavelength range have been widely used for automated processing because they are relatively inexpensive and fast and designed to represent the human eye to detect external features such as size, shape, colour, or the presence of damages and diseases. However, traditional cameras cannot identify damage that is internal or non-visible to the human eye. Hyperspectral imaging combines conventional two-dimensional imaging systems with spectroscopy to simultaneously obtain spatial and spectral information from an object. This technology is being progressively applied for postharvest inspection of fruits and vegetables to develop predictive models for estimating internal quality or identifying key wavelengths related to specific problems. Nevertheless, to obtain accurate and trustable detection, the use of this technology requires adequate equipment, robust calibrations, and powerful statistical tools to create prediction or classification models that are consistent over time. This chapter reviews the application, advantages, and requirements of visible and near-infrared hyperspectral imaging for mechanical/bruise damage and quality assessment of fresh horticultural produce.