Articles
NONDESTRUCTIVE PREDICTION OF INTERNAL BROWNING IN PINEAPPLE USING TRANSMITTANCE SHORT WAVELENGTH NEAR INFRARED SPECTROSCOPY
Article number
989_54
Pages
395 – 399
Language
English
Abstract
Pineapple [Ananas comosus (L.) Merr.] is one of the most important
commercial fruit of Thailand.
The taste and consistency of the fruit is of great
importance, however internal browning, a common physiological disorder
affecting the fruit, which cannot be identified by visual inspection, makes the
product unacceptable for export.
In this study, Near Infrared (NIR) spectroscopy in
the range of 665-955 nm was investigated as a non-destructive means to identify
internal browning.
Partial least squares-discriminant analysis (PLS-DA) was used in
conjunction with the pre-treated NIR spectra as a first step in the development of an
automated method of pineapple fruit sorting.
A set of 243 samples was used for this
research (131 commercially acceptable pineapples and 112 pineapples suffering
from internal browning). A sample of 145 fruits was used for a training set and 98
samples for a test set.
The smoothing and the first derivative pretreatment of
averaged spectra were performed to obtain the best calibration model.
The overall
classification accuracy of the PLS-DA/NIR model on the prediction set was 90.8%
(47 out of 53 for the sound pineapples and 42 out of 45 for the internally browned
pineapples). This study demonstrates that NIR transmittance spectroscopy is
potentially a useful nondestructive method that can be used to predict internal
browning disorder in intact pineapples.
commercial fruit of Thailand.
The taste and consistency of the fruit is of great
importance, however internal browning, a common physiological disorder
affecting the fruit, which cannot be identified by visual inspection, makes the
product unacceptable for export.
In this study, Near Infrared (NIR) spectroscopy in
the range of 665-955 nm was investigated as a non-destructive means to identify
internal browning.
Partial least squares-discriminant analysis (PLS-DA) was used in
conjunction with the pre-treated NIR spectra as a first step in the development of an
automated method of pineapple fruit sorting.
A set of 243 samples was used for this
research (131 commercially acceptable pineapples and 112 pineapples suffering
from internal browning). A sample of 145 fruits was used for a training set and 98
samples for a test set.
The smoothing and the first derivative pretreatment of
averaged spectra were performed to obtain the best calibration model.
The overall
classification accuracy of the PLS-DA/NIR model on the prediction set was 90.8%
(47 out of 53 for the sound pineapples and 42 out of 45 for the internally browned
pineapples). This study demonstrates that NIR transmittance spectroscopy is
potentially a useful nondestructive method that can be used to predict internal
browning disorder in intact pineapples.
Authors
S. Sukwanit, S. Teerachaichayut
Keywords
pineapple, NIR, internal browning, nondestructive
Online Articles (54)