Bruise detection using NIR hyperspectral imaging for strawberry (Fragaria × ananassa Duch.)

Abstract

Bruising in fruits and its detection is a great concern for food safety to the consumers, and for incurrence of economic losses to industry. This research was aimed at developing techniques for detection of compression bruises in 'Akihime' variety strawberries (Fragaria × ananassa Duch.) using near infrared (NIR) hyperspectral imaging. Using 120 fruit samples at 70-80% ripe and full-ripe levels of maturity that had received six levels of bruising force (0 N, 0.5 N, 1.0 N, 1.5 N, 2.0 N and 3.0 N), spectral images were taken from 650 to 1,000 nm at 5 nm wavelength intervals and from 0 to 4 days after bruising, Using stepwise linear discriminant analysis, optimal wavelengths of 825 nm and 980 nm were identified. The three judgment methods (linear discriminant analysis, normalized difference and artificial neural network) had performed equally well, while the normalized difference method was found to be more useful. Additionally, the extent of bruising damage detected in strawberries by the judgment method had markedly decreased with time, which elucidated temporal changes in bruising pattern while in storage.