Distinguishing Carrot’s Characteristics by Near Infrared (NIR) Reflectance and Multivariate Data Analysis

Nawaf Abu-Khalaf, Bent S Bennedsen, Gitte K Bjørn

Abstract


This study presents an attempt to predict the sensory quality of carrots using Near
Infrared (NIR) technology and multivariate data analysis (e.g. PCA, PLS-DA) to
analyse some of factors modulating carrot sensory quality using classification system.
The aim is to introduce a technology to develop a sensor that can non-destructively
predict the sensory quality of different commodities. Such sensors could be used online
in warehouses and public fruit markets. A NIR spectrometer with Photo Diode Array
(PDA) detector, was able to classify different carrot samples according to their cultivar
and production system (organic, conventional) with a high accuracy. Non-destructive
classification and modelling was based on optical reflectance in NIR range (700-1100
nm). Different cultivars and different production system samples were correctly
classified in rates of > 64% and > 82% respectively. The ability of NIR to classify
samples with different cultivars and production systems indicates that NIR is able to
predict the sensory quality of carrots, which are mainly dependant on cultivar,
production system and year.

Full Text:

PDF