Sensing taste of fruits and vegetables using near infrared (NIR) technology

Abstract

This project concerns the feasibility of using near infrared (NIR) technology for sensing taste of fruits and vegetables. This study was conducted based on the known non-destructive ability of NIR to assess some internal quality parameters of fruits and vegetables, like soluble solid contents (SSC), acidity and other chemical components. Ongoing work will evaluate the possibility of designing a NIR taste sensor. This sensor can be used online in fruit grading (according to their taste quality) in warehouses and public fruit markets. Taste of any fruit is considered to be composed from SSC, acidity and characteristic taste parameters of the fruit (distinguished chemicals). The possibility of distinguishing between two different varieties (of the same fruit) having the same SSC/acidity values means that SSC and acidity are eliminated leaving only the characteristic taste parameters in both varieties to be detected. A NIR spectrometer with photo diode array (PDA) detector and fiber optics was used for spectra measurement. Non-destructive prediction and classification was based on optical reflectance in NIR range (700-1100 nm). Three experiments were carried out, one for plums and two for apples. Two different varieties of same fruit were used in each experiment. A reasonable correlation between NIR reflectance and a number of quality parameters was achieved. Also, it was found that it is possible to classify two varieties of fruits. Furthermore, NIR spectroscopy could classify two varieties even when they had the same ratio of SSC and acidity (the verification of the study). Two varieties of plums (Reine Claude and Blackamber), which had the same ratio of SSC and acidity, were both correctly classified in 92.8% of the cases at 5% significant level. Golden Delicious and Jonagold apples were correctly classified in 90% and 83.4% of the cases respectively at 5% significant level. Also, Aroma and Elstar apple samples, which had the same ratio of SSC and acidity, were correctly classified in 100% and 85% cases respectively at 10% significant level. Thus, it can be concluded that NIR technology has a high potential in sensory science for sensing taste of fruits.