Research Progress on Detection of Apple Watercore Based on Visible and Near-Infrared Spectroscopy

Abstract

Apples are one of the most widely produced fruits globally, recognized for their crisp texture, juiciness, and nutritional value. Apples affected by watercore are particularly favored by consumers for their high sugar content and unique flavor. However, during prolonged storage, watercore apples often experience metabolic disorders, making it necessary to develop a rapid, high-throughput, and effective nondestructive testing method to monitor this condition. Near-infrared (NIR) spectroscopy has gained extensive application in apple quality assessment due to its speed, low cost, and ability to measure multiple indices simultaneously. This paper reviews physiological diseases affecting apples, particularly watercore, and discusses various nondestructive testing methods. It emphasizes the current application of visible/near-infrared (Vis/NIR) spectroscopy in detecting watercore in apples. Additionally, this paper addresses the challenges and prospects of using Vis/NIR spectroscopy for watercore detection. This review is aimed at providing insights into more effective ways to manage physiological changes in apples, such as watercore.

Full text

Review Article
Research Progress on Detection of Apple Watercore Based on
Visible and Near-Infrared Spectroscopy
Tao Hu and Qingshen Sun
Engineering Research Center of Agricultural Microbiology Technology, Ministry of Education & Heilongjiang Provincial Key
Laboratory of Plant Genetic Engineering and Biological Fermentation Engineering for Cold Region & Key Laboratory of
Microbiology, College of Heilongjiang Province & School of Life Sciences, Heilongjiang University, Harbin, China
Correspondence should be addressed to Qingshen Sun; sunqingshen@hlju.edu.cn
Received 10 August 2024; Accepted 24 January 2025
Academic Editor: Fabiano A.N. Fernandes
Copyright © 2025 Tao Hu and Qingshen Sun. Journal of Food Processing and Preservation published by John Wiley & Sons Ltd.
This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and
reproduction in any medium, provided the original work is properly cited.
Apples are one of the most widely produced fruits globally, recognized for their crisp texture, juiciness, and nutritional value.
Apples affected by watercore are particularly favored by consumers for their high sugar content and unique flavor. However,
during prolonged storage, watercore apples often experience metabolic disorders, making it necessary to develop a rapid, high-
throughput, and effective nondestructive testing method to monitor this condition. Near-infrared (NIR) spectroscopy has
gained extensive application in apple quality assessment due to its speed, low cost, and ability to measure multiple indices
simultaneously. This paper reviews physiological diseases affecting apples, particularly watercore, and discusses various
nondestructive testing methods. It emphasizes the current application of visible/near-infrared (Vis/NIR) spectroscopy in
detecting watercore in apples. Additionally, this paper addresses the challenges and prospects of using Vis/NIR spectroscopy
for watercore detection. This review is aimed at providing insights into more effective ways to manage physiological changes in
apples, such as watercore.
Keywords: apple quality; near-infrared spectroscopy; nondestructive testing; watercore
1. Introduction
Apples, belonging to the Rosaceae family and the Malus
genus, are renowned for their crisp texture, juiciness, and
nutritional value, making them one of the most abundant
fruits globally [1]. They contain water, fructose (primarily),
glucose, sucrose, organic acids (0.2%–0.8%), dietary fiber
(approximately 2%–3%, with pectin accounting for less than
50% of apple fiber), anthocyanins, polyphenols, and vitamins
(primarily Vitamin C, 2.3–31.1 mg/100 g dry matter base)
[2, 3]. These nutrients provide various health benefits includ-
ing energy supply, digestive support, immune enhancement,
and antioxidant properties [4–6].
Apples can be consumed fresh or used in making bever-
ages and vinegar, contributing to their popularity among
consumers [7, 8]. In recent years, the planting area of apples
in China has been expanding annually [9], and the quality
has also been improving, which is closely related to advanced
management practices, including both preharvest and post-
harvest management [10]. Concurrently, consumers’demand
for apple quality has also been increasing [11–13], with a
growing focus not only on external characteristics such as
color, shape, and size but also on internal qualities like flavor,
texture, and nutritional value [14].
However, due to sociopolitical changes (such as increased
health and environmental awareness), lifestyle shifts, and
evolving fashion trends, consumers’demand for apple quality
varies [15]. For example, while watercore apples are consid-
ered a sign of full maturity and are regarded as premium in
Asia [16], they are often seen as a physiological disorder
in Europe and America. During long-term storage, mild
watercore apples may see their sugar cores disappear, while
severe watercore apples are highly susceptible to browning,
resulting in a loss of edible and commercial value. There-
fore, dynamic monitoring of watercore apples is particularly
important [17].
Wiley
Journal of Food Processing and Preservation
Volume 2025, Article ID 4394346, 13 pages
https://doi.org/10.1155/jfpp/4394346

Recently, visible and near-infrared spectroscopy (Vis-
NIRS) technology has been widely applied in apple quality
detection as a rapid, high-throughput, simple, and nonde-
structive testing method, achieving significant advances
[18, 19]. By studying the physiological disorders of water-
core apples during long-term storage and their rapid detec-
tion methods, it is possible to promptly identify apples
with disappearing sugar cores and internal browning (IB).
This can prevent the loss of commercial value due to sugar
core disappearance and the widespread quality issues caused
by the rotting of internally browned apples infecting healthy
apples. Moreover, compared to traditional manual selection
or destructive testing methods, Vis-NIRS technology signif-
icantly improves the detection speed and efficiency of water-
core apples.
To date, research on the physiological disorders of
watercore apples during long-term storage and their rapid
detection methods has been mainly concentrated in Asia.
However, globally, such research is not yet extensive and
has not formed a systematic framework. This paper summa-
rizes the latest research findings on early detection of physi-
ological disorders caused by watercore in apples during
long-term storage using Vis/NIR spectroscopy technology,
providing valuable references and insights to promote the
development of nondestructive testing techniques for water-
core in apples.
2. Impact of Watercore on Apple Quality
Watercore is a physiological disorder in apples characterized
by the accumulation of sugars, such as sucrose and sorbitol,
in the intercellular spaces of the fruit flesh [20]. The typical
visual feature of watercore is the formation of semitranspar-
ent tissues near the core [21]. Several factors contribute to
this disorder, including liquid flow rich in sorbitol in vascu-
lar tissues (specifically within the sieve element–companion
cell complex (SE-CC)), temperature fluctuations, fruit matu-
rity, and calcium content [22–24]. Different apple varieties
have varying susceptibilities to watercore, with Fuji apples
being particularly prone to this condition [17].
Under normal metabolic conditions, sorbitol converts to
fructose. However, in watercore apples, environmental fac-
tors such as extreme temperatures, significant day–night
temperature fluctuations, high nitrogen content, and an ele-
vated source–sink ratio hinder this conversion. As a result,
sorbitol metabolism is disrupted in cortical cells, and sorbi-
tol continues to accumulate excessively in the phloem cells,
leading to water accumulation in the base of the fruit [25].
Sorbitol, being an osmotic active substance, can attract sub-
stantial amounts of water, causing the flesh tissue to appear
watery and transparent [26]. Typically, apples with high cal-
cium content exhibit a lower incidence of watercore [27],
although this relationship is not always consistent [28]. Fur-
thermore, early-harvested apples tend to be less mature,
smaller, and underdeveloped; therefore, they generally have
a lower incidence of watercore [29].
Interestingly, the accumulation of sugars such as sorbitol
in watercore apples results in a higher sweetness level com-
pared to ordinary apples [30]. While watercore apples main-
tain the same high nutritional value as regular apples, being
rich in vitamins, amino acids, minerals, dietary fiber, and
other bioactive substances, their unique flavor profile
becomes prominent when fully ripe [31]. This distinct char-
acteristic is highly favored by Asian consumers, who recog-
nize the semitransparent appearance of watercore apples as
a symbol of quality [32].
In short, consumer perceptions of watercore apples differ
significantly by region. However, based on postharvest phys-
iological characteristics, watercore apples are generally not
suitable for prolonged storage. In the following sections, we
will discuss some additional issues associated with watercore
apples.
3. Existing Issues of Watercore Apples in
Postharvest Storage
When fruits are harvested from trees, their external nutrient
supply ceases, leading to a gradual decline in quality as the
preharvest accumulated nutrients are utilized to maintain
physiological activities during storage [33]. To mitigate post-
harvest losses and extend the marketable window, apples
need to be stored for several months at around 0
°
C. Gener-
ally, apples have good storage capabilities and can be kept
for extended periods under appropriate conditions [34].
However, apples affected by watercore present challenges
for long-term storage, as their quality deteriorates rapidly.
Mildly affected watercore apples gradually lose their
sugar core, while those severely impacted develop IB [35].
This condition causes the apples to lose both edibility and
commercial value [36]. IB is considered a “serious defect”;
if more than 2% of apples in a batch exhibit this condition,
the entire batch may be rejected [37]. Such defects can occur
under various storage methods, including refrigerated stor-
age (RS), controlled atmosphere (CA), initial low oxygen
stress (ILOS), ultralow oxygen (ULO), and dynamic con-
trolled atmosphere (DCA) [23, 34].
Interestingly, symptoms of mildly affected watercore
apples often diminish during low-temperature storage periods
[38]. This improvement may result from reduced respiration
and relatively stable metabolic activity in low-temperature
conditions, which gradually restores normal sorbitol metabo-
lism and reduces its accumulation in intercellular spaces
[39]. Low-temperature storage also slows the accumulation
of liquid in these spaces, helping to maintain the integrity
and stability of cell walls and membranes. Furthermore, lower
temperatures decrease cell membrane permeability, prevent-
ing excessive liquid from entering intercellular spaces and
thereby mitigating the severity of watercore [40]. Additionally,
low-temperature storage may activate antioxidant defense
mechanisms within the fruit, reducing oxidative stress and
protecting cell structures from damage and liquid accumula-
tion caused by oxidative stress [17, 41, 42].
Despite these potential benefits, watercore significantly
impacts apples during long-term storage, with water-
soaked areas progressively evolving into serious IB [43].
Analysis reveals that sorbitol diffuses from the vascular bun-
dles and, in severe cases, can extend to the fruit surface [44].
The occurrence of IB in apples is linked to the accumulation
2 Journal of Food Processing and Preservation

of ethanol and acetaldehyde in the fruit tissues, as well as
reduced gas diffusion and sensitivity to low oxygen levels
and high carbon dioxide concentrations [45].
Watercore primarily affects the core region of apples,
making it generally imperceptible to the naked eye unless
the condition is extremely severe [46]. Therefore, there is a
pressing need for a nondestructive, reliable, and rapid
method to dynamically monitor watercore in postharvest
apples. Such methods would enable early detection and
timely adjustments, ultimately enhancing both the economic
value and quality of apples.
4. Existing Detection Methods for
Apple Watercore
Currently, nondestructive detection technologies for apple
watercore include methods such as x-ray computed tomog-
raphy (CT), x-ray imaging, thermal imaging, optical density
measurements, and near-infrared (NIR) spectroscopy. These
technologies have garnered significant attention from
researchers both domestically and internationally [47], dem-
onstrating notable achievements in identifying IB and other
defects associated with watercore [48] (Table 1).
4.1. X-Ray Imaging and CT. The essence of IB lies in voids
formed because of cell damage caused by storage pressure
after intracellular fluid fills the cells. These damages are typ-
ically concentrated around the core region of the apple.
Using CT scanning technology, the three-dimensional (3D)
distribution of individual cells and their internal air net-
works within apple tissue affected by IB can be clearly
observed. This method not only identifies the presence of
IB accurately but also quantitatively evaluates the severity
of the defect [55, 58].
Tempelaere et al. utilized x-ray CT and x-ray imaging
technology to obtain 3D and two-dimensional (2D) images
of apples. The study classified experimental data into three
tasks: Task 1—healthy versus defect fruit with anoxic damage
due to N
2
,Task2—healthy versus defect fruit with hypoxic
damage due to CO
2
,andTask3—healthy versus defect fruit
with either anoxic or hypoxic damage. Multiple binary 3D
and 2D ResNet classifiers were employed to detect IB in the
samples for each task, and the BraeNet model was developed
to enhance detection performance: Task1-3D-BraeNet-
N
2
—healthy samples: 160/N
2
-affected samples: 240; Task1-
2D-BraeNet-N
2
—healthy samples: 6400/N
2
-affected samples:
9600; Task2-3D-BraeNet-CO
2
—healthy samples: 535/CO
2
-
affected samples: 840; Task2-2D-BraeNet-CO
2
—healthy sam-
ples: 21,400/CO
2
-affected samples: 33,600; Task3-3D Brae-
Net-N
2
/CO
2
—healthy samples: 695/N
2
-affected samples:
240/CO
2
-affected samples: 840; Task3-2D BraeNet-N
2
/
CO
2
—healthy samples: 28,800/N
2
-affected samples: 9600/
CO
2
-affected samples: 33,600. The 3D BraeNet-N
2
/CO
2
model achieved a maximum detection accuracy of 96%±1%
for N
2
- and CO
2
-affected fruits. In comparison, the 2D Brae-
Net-N
2
and 2D BraeNet-N
2
/CO
2
models achieved maximum
detection accuracies of 99 0%±03%and 96%±1%,respec-
tively, for N
2
-andCO
2
-affected fruits [57].
4.2. Nuclear Magnetic Resonance (NMR). Chayaprasert et al.
utilized a 5.55-MHz low-field NMR system to detect healthy
apples and apples with IB. They observed that the NMR
signal decay rate in apples with IB was slower than that of
the healthy apples, demonstrating the feasibility of NMR
technology in detecting IB in apples [59]. Wang et al. devel-
oped a nondestructive online disease identification program
based on NMR relaxation measurements. They employed a
fast low-angle radiofrequency pulse sequence (FLASH) to
acquire T∗2-weighted and proton density (PD)–weighted
images. Statistical features such as coefficient of variation,
skewness, and kurtosis were extracted from image intensity
histograms for classification. Results showed that PD-
weighted images achieved the best detection performance,
with a classification accuracy of 96% and a detection rate
of 30 samples per minute [60].
4.3. Thermal Imaging. Baranowski et al. utilized dynamic
thermography to differentiate apples with and without
watercore. The experimental sample consisted of 35 water-
core apples and 35 healthy apples. The study employed a
VIGOcam v50 thermal imaging device and two 500-W hal-
ogen lamps as heat sources. The rationale for this study is
that the accumulation of water in the intercellular spaces
of watercore tissue not only increases the mass density of
the fruit but also raises its heat capacity while reducing its
thermal conductivity. Thermal image sequences were cap-
tured at a frequency of 15 frames per second, with approxi-
mately 600 thermal images in each sequence. Compared
with healthy apples, watercore apples exhibited a higher
density range. Using apple density as a classification crite-
rion, the coefficient of determination (R2) for healthy and
watercore apples was 0.6813 and 0.8028, respectively [52].
4.4. Optical Density Measurements. The optical density
method has also proven effective for detecting IB in apples.
Throop et al. quantified watercore damage by measuring
light transmission along the apple’s stem/calyx axis. The
experiment included two sample groups: Group A consisting
of 80 apples, with data collected using a high-sensitivity sil-
icon target vidicon camera, and Group B consisting of 94
apples, with data collected using a CCD monochrome cam-
era. After capturing the image information, the apples were
halved along the equatorial line and visually classified into
five categories, with category one representing healthy
apples. The recognition rates for categories one through five
were 99%, 95%, 95%, 95%, and 99%, respectively. However,
this technique requires precise alignment of the apple’s
calyx, making it less suitable for large-scale online test-
ing [50].
4.5. Challenges and Considerations. While these technologies
each have their own advantages in detecting watercore and
related defects, they exhibit significant differences in terms
of cost-effectiveness, ease of use, and detection accuracy
compared to NIR spectroscopy [55, 58] (Table 1). X-ray,
CT, MRI, and NMR offer high resolution and precise identi-
fication capabilities but come with high equipment costs,
complex operation, and time-consuming processes, posing
3Journal of Food Processing and Preservation

a higher entry barrier [61]. Thermal imaging and optical
density measurement are moderately priced compared to
NIR spectroscopy but fall short in terms of resolution and
operational flexibility [52]. In contrast, NIR spectroscopy
stands out as the ideal choice for detecting watercore and
related defects due to its fast detection speed, high efficiency,
ease of operation, and ability to perform real-time online
detection. These advantages make it highly suitable for
large-scale commercial applications, showcasing significant
potential and promising prospects.
5. Application Status of Vis/NIR Spectroscopy
Technology in Apple Quality Detection
In recent years, Vis/NIR spectroscopy technology has gained
widespread use for the rapid and nondestructive assessment
of apple quality [62] (Table 2). Compared to other nonde-
structive detection methods such as MRI, x-ray, and CT,
Vis/NIR spectroscopy offers several advantages, including
high detection efficiency, simple sample preparation, low
cost, and the ability to analyze multiple components simul-
taneously [52]. A schematic diagram of the spectroscopic
acquisition system is shown in Figure 1.
5.1. Detection of Watercore in Apples. Zhang et al. conducted
an online analysis of watercore apples using Vis/NIR full-
transmission spectroscopy (680–1000 nm). They achieved
classification success rates of 100% for healthy apples and
96.87% for watercore apples. Their models, based on effec-
tive wavelengths, utilized partial least squares (PLS), multi-
ple linear regression (MLR), and least squares support
vector machine (LS-SVM) techniques to predict the severity
of apple watercore [67]. Similarly, Guo et al. developed a
PLS linear regression model that successfully predicted
watercore severity (RP = 0 943) and explored the potential
of combining coherent anti-Stokes Raman scattering with
convolutional neural network (CARS-CNN) methods [69].
Additionally, researchers established quantitative detection
models for apple watercore using NIR spectroscopy. For
instance, the CARS-PLS model demonstrated excellent pre-
dictive performance (RP = 0 9562, root mean square error
of prediction RMSEP =1340%), highlighting the potential
of this technology in watercore detection [66]. Li et al.
successfully developed an evaluation model for Fuji apple
watercore using online Vis-NIRS technology, demonstrating
the application prospects of this technology in real-time
detection [68].
5.2. Detection of IB. McGlone et al. utilized NIR transmis-
sion technology to detect IB in apples online, constructing
a dynamic high-speed fruit measurement system (LAS).
Their measurement results showed an R2of approximately
0.9 and a RMSEP of about 4.1%, demonstrating the system’s
effectiveness in identifying IB in apples [63]. Sun et al. used
NIR measurement devices with two different optical geome-
tries to detect vascular browning (VAB) in apples. Using
Table 1: Detection of watercore disease and internal browning of apple by different nondestructive testing techniques.
Time Method Application Advantages Limitations Online
detection References
1988 Nuclear magnetic
resonance (NMR) Internal browning
•High resolution
(0 43 × 0 43 mm)
•Nondestructive
•High cost
•Complex operation
•Detection speed (8.4 min/apple)
Support [49]
1994 Optical density Watercore
•Detection speed
(2 s/apple)
•Nondestructive
•Subject to environmental influences
•Low resolution (360 V × 272H)Nonsupport [50]
1998 MRI Watercore
•High resolution
(390 × 390 μm)
•Nondestructive
•High cost
•Complex operation
•Detection speed (25 min/apple)
Support [51]
2008 Thermography Watercore
•Detection speed
(40 s/apple)
•Nondestructive
•Low resolution (384 × 288 mm)
•Subject to environmental influences Nonsupport [52]
2009 Thermography Watercore
•Detection speed
(40 s/apple)
•Nondestructive
•Low resolution (384 × 288 mm)
•Subject to environmental influences Nonsupport [53]
2013 MRI Internal browning
•High resolution
(2048 × 2048 mm)
•Nondestructive
•High cost
•Detection speed (20 min/apple)
•Complex operation
Support [54]
2021 X-ray Internal browning
•High resolution
(voxel 3 μm)
•Nondestructive
•High cost
•Detection speed (45 min/apple)
•Complex operation
Support [55]
2024 CT Internal browning
•High resolution
(voxel 129.3 μm)
•Nondestructive
•High cost
•Complex operation
•Detection speed (2.4 min/apple)
Support [56]
2024 X-ray Internal browning •Strong penetration
•Nondestructive •High resolution Support [57]
4 Journal of Food Processing and Preservation

partial least squares discriminant analysis (PLS-DA), they
developed classification models for the measurement data
under various conditions. When the NIR measurement
device (System 2) was optimally positioned at 90
°
and a
detection threshold was set to identify 80% of defective
apples, 21% of healthy apples were misclassified. This simu-
lation demonstrated the feasibility of NIR spectroscopy sys-
tems for detecting VAB, though it indicated limitations in
sample volume detection [65]. Wang et al. also employed
NIR spectroscopy to detect IB in apples, establishing dis-
criminant models using peak area discriminant analysis
(PADA), principal component analysis discriminant analy-
sis (PCADA), and partial PLS-DA. All three methods
achieved a correct discrimination rate of 100% for IB
apples, showcasing the rapid and nondestructive identifica-
tion capabilities of NIR spectroscopy [64]. Additionally,
Mogollon et al. used NIR spectroscopy to detect IB occur-
ring in “Cripps Pink”apples during cold storage. They
developed a quantitative support vector machine regression
(SVMR) model to predict the percentage of apple IB area,
achieving an R2of 0.70 after 90 days of storage, with root
mean square error of calibration (RMSEC) and RMSEP
datasets at 18% and 15%, respectively, and a misclassifica-
tion rate of 12% [70].
Overall, NIR spectroscopy technology demonstrates
strong performance in detecting physiological disorders
such as watercore degradation and IB in apples during
long-term storage. This includes the identification of apple
watercore [67], assessment of apple watercore severity [67,
69], detection of IB [63] and apple VAB [65], and determi-
nation of IB severity [70]. By enabling dynamic monitoring
of apple physiological disorders during storage, Vis/NIR
spectroscopy facilitates early detection and timely adjust-
ments to maintain apple quality.
Table 2: Detection of watercore and internal browning of apple based on Vis/NIR spectroscopy.
Time Method Application Sample Sample
division Algorithms R2Accuracy
(%) References
2015 NIR Internal
browning 87/30 PLS 0.9 [63]
2008 Vis/NIR Watercore Healthy 334, fall ill 240 PLSDA 96.7% [64]
Internal
browning Healthy 334, fall ill 62 PLSDA 100%
2018 NIR Internal
browning
Healthy 38, mild 27,
severe 16 PLSDA 79% [65]
2020 Vis/NIR Watercore Healthy 52, mild 97,
moderate 74, severe 95 238/80 kNN 95% [17]
2020 NIR Watercore Healthy 52, fall ill 107 100/59 PLS 0.96 [66]
2022 Vis/NIR Watercore Healthy 138, fall ill 127 199/66 PLS-DA 100% [67]
LS-SVM 96.87%
2022 Vis/NIR Internal
browning Healthy 100, fall ill 100 200/66 CARS-PLS-
DA 89.39% [62]
2023 Vis/NIR Watercore Healthy 138, fall ill 127 199/66 MC-UVE-
SPA-PLS-DA 95.96% [68]
Spectrometer
Position sensor
Dark box
Automatic reference
Halogen lamp
Fruit
Fruit cup
Belt conyeyor
Computer
Data cable
Optical fiber
Spectrometer
Figure 1: Diagram of Vis/NIR spectroscopy measurement system.
5Journal of Food Processing and Preservation

6. Application of Vis/NIR Spectroscopy in
Online Detection of Internal
Defects in Fruits
Internal defects in fruits (moldy core and IB) greatly impact
their market value and consumer acceptance [71]. Online
detection of these internal defects not only improves sorting
efficiency but also reduces production waste and ensures the
consistency and safety of fruit quality. Vis-NIRS, with its
nondestructive and efficient detection capabilities, has been
extensively studied for the online detection of internal
defects in fruits, as traditional equipment often falls short
of practical requirements due to its large size, immobility,
and stringent environmental conditions.
6.1. Online NIR Spectroscopy Detection System. For online
NIR spectroscopy detection system, Hu et al. developed a
system for the online detection of moldy core in apples, aim-
ing to reduce the impact of apple size and soluble solid con-
tent (SSC) on detection results and achieve fast and accurate
moldy core identification. The system significantly improved
detection performance by applying secondary spectral cor-
rection to SSC features. At a running speed of 0.14 m/s, the
model was evaluated using 220 training samples and 50 test
samples, achieving detection accuracies of 94.44% and
88.33% on the training and test sets, respectively [72].
Another work by Tian et al. employed a short integration
time mode to collect full-transmission spectra of apples
online at a speed of 0.5 m/s in three different orientations:
T1 (stem–calyx axis horizontal and perpendicular to the
conveyor belt), T2 (stem–calyx axis horizontal and parallel
to the conveyor belt), and T3 (stem–calyx axis horizontal
and intersecting the conveyor belt). The study revealed that
spectral collection in the T2 orientation was more suitable
for apple moldy core detection. Based on spectral data col-
lected in the T2 orientation, an optimal classification model
was developed using a total of 279 samples (187 healthy and
92 affected), with the training and test sets split in a 3:1 ratio.
In the test set, the model achieved classification accuracies of
86.9% for healthy samples and 93.9% for apple moldy core
samples [73].
Zhang et al. developed an online nondestructive detec-
tion system for apple moldy core in apples using NIR trans-
mission spectroscopy (600–1100 nm). The study involve d
120 apple samples, which were randomly divided into train-
ing and test sets in a 2:1 ratio. Spectral data were collected
from apples in three different orientations: T1 (stem–calyx
axis horizontal and perpendicular to the conveyor belt), T2
(stem–calyx axis horizontal and parallel to the conveyor
belt), and T3 (stem–calyx axis horizontal and intersecting
the conveyor belt). The results indicated that a multispectral
model using the average spectra from all three orientations
yielded the best performance. Based on the sequential pro-
jection algorithm (SPA), the multispectral model achieved
classification accuracies of 96.7%, 97.5%, and 97.5% for the
three orientations in the test set, respectively [74].
Guo et al. successfully achieved efficient transfer of NIR
spectroscopy detection models across different devices and
batches by building a cloud data platform and employing
deep learning models. The study utilized 200 apple samples
to collect NIR spectral data and measure SSC. The detection
system consisted of a handheld NIR spectrometer, an Inter-
net of Things (IoT) module, and a cloud data platform. After
spectral data were uploaded to the cloud, an autoencoder
(AE) neural network was applied for spectral correction
and model transfer. The results demonstrated that the AE
model significantly reduced differences between detection
terminals and batches, achieving efficient model transfer.
Furthermore, the average transmission time for detection
results was 1.5–2 s, with a transmission efficiency of 100%.
This study provides new technical directions and practical
support for the application of handheld spectrometers in
online fruit detection [75].
6.2. Handheld NIR Spectrometers. For handheld NIR spec-
trometers, Pissard et al. investigated the analytical perfor-
mance of a handheld NIR spectrometer (Micro) compared
to a benchtop NIR spectrometer (XDS) and explored the fea-
sibility of transferring calibration models between the two
devices, revealing that the handheld spectrometer Micro per-
formed comparably to the benchtop spectrometer XDS in
detecting apple quality attributes, including SSC, titratable
acidity (TA), flesh firmness (PF), and starch–iodine index
(SII). Additionally, using the direct standardization (DS)
method, successful calibration model transfer between the
devices was achieved, providing new technical directions
and practical support for the application of handheld spec-
trometers in online detection [76].
In addition, companies in Japan (such as Shibuya, Mit-
sui) and China (such as Wuxi Xunjie Guangyuan) have
introduced corresponding NIR online detection devices
and multiple handheld NIR spectrometers, representing the
Japanese Shibuya Optical VGA-InGaAs-NVU3VD and the
Chinese Xunjie Guangyuan IAS-OnLine S100, respectively.
The NIR online detection devices, equipped with built-in
image processing chips, are renowned for their high speed
and high resolution. They provide strong technical support
for modern agricultural quality management and grading
and sorting. Future research in NIR spectroscopy will focus
on miniaturization, cost reduction, low power consumption,
and enhanced portability of devices [77].
7. Analysis and Discussion
7.1. Identification and Severity Assessment of Apple
Watercore. Apple watercore is a physiological disorder that
affects the interior of apples. Identifying this condition typi-
cally involves collecting spectral data from three different
detection orientations relative to the conveyor belt: parallel,
intersecting, and perpendicular to the calyx axis (as illus-
trated in Figure 2). To minimize positional effects within
the apple fruit, each apple is scanned three times (see
Figure 3) to obtain averaged spectra [67–69, 78].
The raw spectra of watercore apples and healthy apples
show significant differences, primarily due to disruptions
in sorbitol metabolism under environmental conditions such
as large diurnal temperature variations. Specifically, when
the normal metabolism of sorbitol in cortical cells is
6 Journal of Food Processing and Preservation

(a)
(b)
(c)
Figure 2: Three different orientations: (a) apple stem–calyx axis vertical, stem upward; (b) apple stem–calyx axis horizontal and parallel to
the moving direction of the conveyor belt, calyx in front; (c) apple stem–calyx axis horizontal, stem toward light source.
(a) (b)
(c)
Figure 3: Apple rotation diagram. Rotate 120
°
(a), 240
°
(b), and 360
°
(c) on the apple stem–calyx axis.
7Journal of Food Processing and Preservation

inhibited, it accumulates continuously in phloem cells. As an
osmotically active substance, sorbitol can absorb a large
amount of water, resulting in a significant increase in water
content and reduction of the air in the intercellular spaces
in watercore apples as compared to that of the healthy apples
[21]. These physiological changes collectively lead to the dis-
tinct spectral characteristics between watercore and healthy
apples. Watercore apples typically exhibit two prominent
absorption peaks at 720 and 810 nm, but these peaks overlap
significantly with the spectral features of healthy apples,
making it challenging to differentiate using raw spectra
[67]. This spectral difference is also evident in spectra col-
lected in intersecting and perpendicular orientations. How-
ever, spectra collected parallel to the calyx axis demonstrate
higher identification accuracy. This improvement in preci-
sion can be attributed to the fact that the waterlogged glassy
tissue of watercore apples primarily surrounds the core and
the stem–calyx center, enhancing light transmission. Never-
theless, the raw spectra are influenced by noise, significant
scattering, and the overlapping spectral features of healthy
apples, complicating the distinction between watercore and
healthy apples. Therefore, further processing of the raw spec-
tra is necessary to enhance the differentiation.
Initially, the spectral data undergoes preprocessing.
Smoothing via the Savitzky–Golay (SG) filter is typically
applied to reduce spectral random noise [79]. The standard
normal variate (SNV) transformation is then used to elimi-
nate spectral differences caused by scattering effects [80].
To enhance model prediction accuracy, effective wave-
lengths in the spectral data are selected using the multivari-
ate scattering and successive projection algorithm (SPA)
[81]. SPA, a forward selection method, effectively isolates
spectral features relevant to apple watercore by projecting
the raw data into a lower-dimensional subspace while retain-
ing the most representative and informative features. This
approach helps reduce dataset complexity and the impact
of noise, thereby improving model robustness and predic-
tion capability.
Following this, initial wavelengths and their numbers are
determined based on the minimum RMSEP through MLR
calibration on the validation set [82]. Models generally fall
into two categories: classification and regression. Regression
models quantify chemical contents, while classification
models categorize targets into distinct groups. In apple
watercore identification, partial PLS-DA and LS-SVM are
commonly used to establish classification models that distin-
guish between watercored and healthy apples [83, 84].
To assess the severity of apple watercore, complementary
detection methods are required. This includes segmenting
the apple along the transverse section (see Figure 4) to deter-
mine if the tested sample exhibits watercore defects and to
establish models [69, 78]. The original RGB image of the
apple is transformed into a grayscale image, and the apple
area is segmented using a thresholding method to calculate
the number of pixels in the apple region. Since the contrast
between watercore tissue and healthy tissue is more pro-
nounced in the B channel image, the B component is
extracted from the RGB image, and a thresholding method
is applied to segment the watercore tissue. The pixel count
of the watercore area is calculated based on the obtained
binary image. Finally, the ratio of pixels between the water-
core area and the entire fruit area is used as a quantitative
measure to evaluate the severity of watercore in apples [66,
69]. The determination of apple watercore severity belongs
to a quantitative model, typically using PLS, MLR, and LS-
SVM models to predict the severity of apple watercore in
apples [85].
7.2. Identification of IB and Determination of IB Severity.
Detection of IB in apples follows a similar process to water-
core detection, involving the acquisition of raw spectral data,
preprocessing, feature spectrum extraction, and qualitative
or quantitative model analysis [63, 70]. However, the primary
distinction lies in the physicochemical indicators collected
during spectral data acquisition. In early-stage watercored
apples, sorbitol primarily affects vascular bundles, typically
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(a)
(1) (2) (3)
(4) (5) (6)
(7) (8) (9)
(b)
Figure 4: Schematic diagram of apple cutting (a) and cross-sectional view after cutting (b).
8 Journal of Food Processing and Preservation

diffusing from these bundles. Therefore, detecting VAB is
significant for early IB detection [86].
VAB manifests as small voids around or above the vas-
cular bundles, often deep within the apple flesh, and appears
as light brown on the surface. Due to their small size, prox-
imity to the core, and lack of strong absorbance, VAB is
challenging to detect. To overcome this challenge, clinical
MRI imaging systems or x-rays are commonly used to scan
apples and identify defective ones, thereby creating a well-
balanced and efficient NIR spectroscopy analysis dataset
[58, 87]. Visual inspection, based on MRI and surface inci-
sion images, is used as an indicator to assess VAB in water-
core apples [65].
8. Limitations and Prospects of NIR
Spectroscopy in Apple Watercore Detection
8.1. Advantages. Nondestructive testing: NIR spectroscopy
enables noninvasive inspection, preserving the integrity of
apples. This allows for multiple tests at different time points
to monitor changes during storage, making it suitable for
large-scale commercial testing.
Rapid and efficient: The technology offers fast testing
speeds, facilitating the examination of a large number of
samples in a short time. This efficiency is particularly bene-
ficial for quick screening on production lines.
Multi-Index measurement: Beyond detecting watercore,
NIR spectroscopy can measure internal characteristics such
as sugar content, acidity, and firmness. This comprehensive
quality information aids in optimizing pre- and postharvest
storage management.
Cost-effective operation: Compared to chemical analysis
methods, NIR technology requires no expensive reagents
or complex instrument maintenance, resulting in lower
operating costs.
High accuracy: By analyzing spectral data, NIR spectros-
copy can accurately identify watercore and other defects,
providing high accuracy and repeatability in results.
8.2. Limitations. Poor portability: Most existing NIR equip-
ment is benchtop and large, limiting its application in field
environments due to lack of portability.
Complex data processing: The large volumes of data gen-
erated require sophisticated mathematical models and algo-
rithms for processing, which places high demands on
computer performance and processing speed.
Model dependency: The effectiveness of NIR models
relies on standardized calibration and transfer procedures.
Variations in sample characteristics, equipment parameters,
and light source conditions can impact the model’s univer-
sality and accuracy.
Environmental sensitivity: Test results can be signifi-
cantly influenced by environmental factors such as tempera-
ture and humidity, necessitating strict control of the testing
environment.
8.3. Future Development Directions. Lowering equipment
costs: Future research should focus on reducing the commer-
cialization costs of NIR detection equipment. This would
enhance economic viability and promote broader adoption
among fruit growers and processing companies.
Improving detection accuracy and speed: Enhancing the
accuracy and speed of NIR spectroscopy is crucial. This can
be achieved through the development of improved spectral
analysis algorithms and data processing techniques, espe-
cially for real-time monitoring and large-scale fruit detection.
Optimizing calibration processes: The calibration and
modeling of NIR spectrometers are complex yet essential
for ensuring detection accuracy. Specialized knowledge is
required, increasing operational difficulty and maintenance
costs. Regular updates to calibration models are necessary
to accommodate different batches and varieties of apples.
Multiparameter integrated detection: NIR technology
should be expanded to enable the integrated detection of
various apple quality factors, including bruising, mechanical
damage, hardness, and complex physiological and patholog-
ical changes. This would provide a more comprehensive
quality assessment.
Integration of smart systems: Integrating NIR technology
with IoT and artificial intelligence (AI) can lead to intelligent
detection systems. These systems would facilitate auto-
mated, real-time monitoring and data analysis of apples,
allowing for timely detection and management of diseases
during storage and transportation, thereby minimizing
losses.
Development of dedicated sensors: Research should focus
on developing dedicated NIR sensors specifically for apple
watercore detection. These sensors should operate reliably
under various storage conditions and possess strong anti-
interference capabilities to ensure accurate and dependable
detection results.
9. Summary
This paper reviews the physiological disease problem of
watercore in apples during long-term postharvest storage
and the corresponding nondestructive detection methods.
It primarily focuses on the application status of Vis/NIR
technology in detecting apple watercore. The existing
nondestructive testing techniques are comprehensively
classified, comparing their applications and accuracy in
assessing apple watercore in apple quality. NIR spectros-
copy is suitable for identifying watercore in apples and
determining the severity of the disease, as well as recogniz-
ing IB and assessing its extent. In conclusion, Vis/NIR
technology is effective for detecting apple watercore in
apples. During postharvest long-term storage, dynamic
monitoring of apple watercore in apples allows for early
detection of IB and watercore degradation, thus better pre-
serving the quality of affected apples.
Data Availability Statement
Our data are available on reasonable request.
Conflicts of Interest
The authors declare no conflicts of interest.
9Journal of Food Processing and Preservation

Funding
The study was funded by the Natural Science Foundation
of Heilongjiang Province (LH2021C075) and the Basic
Research Business Expenses and Research Projects of Provin-
cial Higher Education Institutions in Heilongjiang Province
(2022-KYYWF-1077).
Acknowledgments
The authors thank Qingshen Sun at Heilongjiang University
for the provision of item support to carry out the study.
References
[1] S. Musacchi and S. Serra, “Apple fruit quality: overview on pre-
harvest factors,”Scientia Horticulturae, vol. 234, pp. 409–430,
2018.
[2] Y. Peng and R. Lu, “Analysis of spatially resolved hyperspectral
scattering images for assessing apple fruit firmness and soluble
solids content,”Postharvest Biology and Technology, vol. 48,
no. 1, pp. 52–62, 2008.
[3] S. Fan, B. Zhang, J. Li, C. Liu, W. Huang, and X. Tian, “Predic-
tion of soluble solids content of apple using the combination of
spectra and textural features of hyperspectral reflectance imag-
ing data,”Postharvest Biology and Technology, vol. 121, pp. 51–
61, 2016.
[4] A. Leccese, S. Bartolini, and R. Viti, “Antioxidant properties of
peel and flesh in ‘GoldRush’and ‘Fiorina’scab-resistant apple
(Malus domestica) cultivars,”New Zealand Journal of Crop
and Horticultural Science, vol. 37, no. 1, pp. 71–78, 2009.
[5] B. M. Nicolaï, K. Beullens, E. Bobelyn et al., “Nondestructive
measurement of fruit and vegetable quality by means of NIR
spectroscopy: a review,”Postharvest Biology and Technology,
vol. 46, no. 2, pp. 99–118, 2007.
[6] A. Tschida, V. Stadlbauer, B. Schwarzinger et al., “Nutrients,
bioactive compounds, and minerals in the juices of 16 varieties
of apple (Malus domestica) harvested in Austria: a four-year
study investigating putative correlations with weather condi-
tions during ripening,”Food Chemistry, vol. 338, Article ID
128065, 2021.
[7] J. Trcek, A. Mahnic, and M. Rupnik, “Diversity of the microbi-
ota involved in wine and organic apple cider submerged vine-
gar production as revealed by DHPLC analysis and next-
generation sequencing,”International Journal of Food Micro-
biology, vol. 223, pp. 57–62, 2016.
[8] Y. Li, T. T. H. Nguyen, J. Jin et al., “Brewing of glucuronic acid-
enriched apple cider with enhanced antioxidant activities
through the co-fermentation of yeast (Saccharomyces cerevi-
siae and Pichia kudriavzevii) and bacteria (lactobacillus plan-
tarum),”Food Science and Biotechnology, vol. 30, no. 4,
pp. 555–564, 2021.
[9] N. Wang, J. Wolf, and F.-S. Zhang, “Towards sustainable
intensification of apple production in China —yield gaps
and nutrient use efficiency in apple farming systems,”Journal
of Integrative Agriculture, vol. 15, no. 4, pp. 716–725, 2016.
[10] R. Chen, C. Zhang, B. Xu et al., “Predicting individual apple
tree yield using UAV multi-source remote sensing data and
ensemble learning,”Computers and Electronics in Agriculture,
vol. 201, Article ID 107275, 2022.
[11] A. Normann, M. Röding, and K. Wendin, “Sustainable fruit
consumption: the influence of color, shape and damage on
consumer sensory perception and liking of different apples,”
Sustainability, vol. 11, no. 17, p. 4626, 2019.
[12] J. Mccluskey, R. C. Mittelhammer, A. B. Marin, and K. S.
Wright, “Effect of quality characteristics on consumers’will-
ingness to pay for gala apples,”Canadian Journal of Agricul-
tural Economics-Revue Canadienne D'Agroeconomie, vol. 55,
no. 2, pp. 217–231, 2007.
[13] F. R. Harker, E. M. Kupferman, A. B. Marin, F. A. Gunson, and
C. M. Triggs, “Eating quality standards for apples based on
consumer preferences,”Postharvest Biology and Technology,
vol. 50, no. 1, pp. 70–78, 2008.
[14] M. C. Kyriacou and Y. Rouphael, “Towards a new definition of
quality for fresh fruits and vegetables,”Scientia Horticulturae,
vol. 234, pp. 463–469, 2018.
[15] M. Schreiner, M. Korn, M. Stenger, L. Holzgreve, and
M. Altmann, “Current understanding and use of quality char-
acteristics of horticulture products,”Scientia Horticulturae,
vol. 163, pp. 63–69, 2013.
[16] S. Kasai and O. Arakawa, “Antioxidant levels in watercore tis-
sue in ‘Fuji’apples during storage,”Postharvest Biology and
Technology, vol. 55, no. 2, pp. 103–107, 2010.
[17] F. Tanaka, F. Hayakawa, and M. Tatsuki, “Flavor and texture
characteristics of ‘Fuji'’and related apple (Malus domestica
L.) cultivars, focusing on the rich watercore,”Molecules,
vol. 25, no. 5, p. 1114, 2020.
[18] H. Chang, Q. Wu, H. Tian, J. Yan, X. Luo, and H. Xu, “Non-
destructive identification of internal watercore in apples based
on online Vis/NIR spectroscopy,”Transactions of the ASABE,
vol. 63, no. 6, pp. 1711–1721, 2020.
[19] V. Cortés, J. Blasco, N. Aleixos, S. Cubero, and P. Talens,
“Monitoring strategies for quality control of agricultural prod-
ucts using visible and near-infrared spectroscopy: a review,”
Trends in Food Science and Technology, vol. 85, pp. 138–148,
2019.
[20] A. Cebulj, M. Mikulic-Petkovsek, C. R. Lucaciu et al., “Alter-
ation of the phenylpropanoid pathway by watercore disorder
in apple (Malus x domestica),”Scientia Horticulturae,
vol. 289, p. 110438, 2021.
[21] Z. Liu, M. Du, H. Liu et al., “Chitosan films incorporating litchi
peel extract and titanium dioxide nanoparticles and their
application as coatings on watercored apples,”Progress in
Organic Coatings, vol. 151, Article ID 106103, 2021.
[22] H. Wang, J. Yuan, T. Liu et al., “Fruit canopy position and har-
vest period affect watercore development and quality of the
‘Fuji’apple cultivar fruit,”Scientia Horticulturae, vol. 311,
Article ID 111793, 2023.
[23] H. J. Kweon, M. J. Kim, Y. S. Moon et al., “Relationship
between preharvest factors and the incidence of storage disor-
ders in ‘Fuji’apples during CA storage,”Korean Journal of
Horticultural Science and Technology, vol. 30, no. 1, pp. 50–
55, 2012.
[24] S. K. Kim, D. G. Choi, and Y. M. Choi, “Relationship between
the temperature characteristics and the occurrence of watercore
at various altitudes in ‘Hongro’and ‘Fuji’apples,”Horticultural
Science and Technology, vol. 41, no. 5, pp. 595–604, 2023.
[25] M. Kunihisa, M. F. Minamikawa, R. Yano et al., “Susceptibility of
apple cultivars to watercore disorder is associated with expression
of bidirectional sugar transporter gene MdSWEET12a,”Scientia
Horticulturae, vol. 334, Article ID 113297, 2024.
[26] B. E. Algul, Y. A. Shoffe, D. Park, W. B. Miller, and C. B. Wat-
kins, “Preharvest 1-methylcyclopropene treatment enhances
10 Journal of Food Processing and Preservation

‘stress-associated watercore’dissipation in ‘Jonagold’apples,”
Postharvest Biology and Technology, vol. 181, Article ID
111689, 2021.
[27] J. X. Guo, Y. J. Ma, Z. M. Guo, H. Huang, Y. Shi, and J. Zhou,
“Watercore identification of Xinjiang Fuji apple based on
manifold learning algorithm and near infrared transmission
spectroscopy,”Spectroscopy and Spectral Analysis, vol. 40,
no. 8, pp. 2415–2420, 2020.
[28] T. Yamane, H. Hayama, N. Mitani, H. Inoue, and S. Kusaba,
“Contribution of several fruit quality factors and mineral ele-
ments to water-soaked brown flesh disorder in peaches,”
Scientia Horticulturae, vol. 272, Article ID 109523, 2020.
[29] N. Kumari, J. N. Sharma, and D. Singh, “Effect of harvest
maturity and pre-cooling on post harvest rots of apple,”Inter-
national Journal of Economic Plants, vol. 6, no. 3, pp. 111–115,
2019.
[30] H. Wada, K. Nakata, H. Nonami et al., “Direct evidence for
dynamics of cell heterogeneity in watercored apples: turgor-
associated metabolic modifications and within-fruit water
potential gradient unveiled by single-cell analyses,”Horticul-
ture Research, vol. 8, no. 1, p. 187, 2021.
[31] M. Kunihisa, S. Moriya, K. Abe et al., “Genomic dissection of a
‘Fuji’apple cultivar: re-sequencing, SNP marker development,
definition of haplotypes, and QTL detection,”Breeding Sci-
ence, vol. 66, no. 4, pp. 499–515, 2016.
[32] M. Yang, Q. Lin, Z. Luo et al., “Ongoings in the apple water-
core: first evidence from proteomic and metabolomic analy-
sis,”Food Chemistry, vol. 402, Article ID 134226, 2023.
[33] Y. Zhao, X. Zhu, Y. Hou, Y. Pan, L. Shi, and X. Li, “Effects of
harvest maturity stage on postharvest quality of winter jujube
(Zizyphus jujuba Mill. cv. Dongzao) fruit during cold storage,”
Scientia Horticulturae, vol. 277, Article ID 109778, 2021.
[34] A. Mditshwa, O. A. Fawole, and U. L. Opara, “Recent develop-
ments on dynamic controlled atmosphere storage of apples-a
review,”Food Packaging and Shelf Life, vol. 16, pp. 59–68,
2018.
[35] W. Li, Z. Liu, H. Wang et al., “Heat shock pretreatment and
low temperature fluctuation cold storage maintains flesh qual-
ity and retards watercore dissipation of watercored ‘Fuji’
apples,”Scientia Horticulturae, vol. 323, Article ID 112492,
2024.
[36] M. J. Du, Z. T. Liu, X. T. Zhang et al., “Effect of pulsed con-
trolled atmosphere with CO
2
on the quality of watercored
apple during storage,”Scientia Horticulturae, vol. 278, Article
ID 109854, 2021.
[37] B. P. Khatiwada, P. P. Subedi, C. Hayes, L. C. C. Carlos, and
K. B. Walsh, “Assessment of internal flesh browning in intact
apple using visible-short wave near infrared spectroscopy,”
Postharvest Biology and Technology, vol. 120, pp. 103–111,
2016.
[38] A. A. Saquet, “Storage of ‘Cox Orange Pippin’apple severely
affected by watercore,”Erwerbs-obstbau, vol. 62, no. 4,
pp. 391–398, 2020.
[39] R. Pedreschi, C. Franck, J. Lammertyn et al., “Metabolic profil-
ing of ‘Conference’pears under low oxygen stress,”Postharvest
Biology and Technology, vol. 51, no. 2, pp. 123–130, 2009.
[40] C. Franck, J. Lammertyn, Q. T. Ho, P. Verboven, B. Verlinden,
and B. A. Nicolaï, “Browning disorders in pear fruit,”Posthar-
vest Biology and Technology, vol. 43, no. 1, pp. 1–13, 2007.
[41] A. Zupan, M. Mikulic-Petkovsek, F. Stampar, and R. Veberic,
“Sugar and phenol content in apple with or without water-
core,”Jounal of the Science Food and Agriculture, vol. 96,
no. 8, pp. 2845–2850, 2016.
[42] Q. T. Ho, P. Verboven, B. E. Verlinden, J. Lammertyn,
S. Vandewalle, and B. M. Nicolai, “A continuum model for
metabolic gas exchange in pear fruit,”PLoS Computational
Biology, vol. 4, no. 3, Article ID e1000023, 2008.
[43] A. Melado-Herreros, M. A. Muñoz-García, A. Blanco, J. Val,
M. E. Fernández-Valle, and P. Barreiro, “Assessment of water-
core development in apples with MRI: effect of fruit location in
the canopy,”Postharvest Biology and Technology, vol. 86,
pp. 125–133, 2013.
[44] J. Hou, Z. He, D. Liu et al., “Mechanical damage characteristics
and nondestructive testing techniques of fruits: a review,”Food
Science and Technology, vol. 43, Article ID e001823, 2023.
[45] X. Li, Z. Liu, Y. Ran et al., “Short-term high oxygen pre-
stimulation inhibits browning of fresh-cut watercored Fuji
apples,”Postharvest Biology and Technology, vol. 191, Article
ID 111959, 2022.
[46] A. Itai, “Watercore in fruits,”in Abiotic stress biology in horti-
cultural plants, Y. Kanayama and A. Kochetov, Eds., pp. 127–
145, Springer Japan, Tokyo, 2015.
[47] A. Melado-Herreros, M. E. Fernández-Valle, and P. Barreiro,
“Non-destructive global and localized 2D T1/T2 NMR relaxo-
metry to resolve microstructure in apples affected by water-
core,”Food and Bioprocess Technology, vol. 8, no. 1, pp. 88–
99, 2015.
[48] E. Herremans, A. Melado-Herreros, T. Defraeye et al., “Com-
parison of X-ray CT and MRI of watercore disorder of differ-
ent apple cultivars,”Postharvest Biology and Technology,
vol. 87, pp. 42–50, 2014.
[49] S. Y. Wang, P. C. Wang, and M. Faust, “Non-destructive detec-
tion of watercore in apple with nuclear magnetic resonance
imaging,”Scientia Horticulturae, vol. 35, no. 3-4, pp. 227–
234, 1988.
[50] J. A. Throop, D. J. Aneshansley, and B. L. Upchurch, “Camera
system effects on detecting watercore in 'Red Delicious'
apples,”Transactions of the ASAE, vol. 37, no. 3, pp. 873–
877, 1994.
[51] C. J. Clark, J. S. Macfall, and R. L. Bieleski, “Loss of watercore
from ‘Fuji’apple observed by magnetic resonance imaging,”
Scientia Horticulturae, vol. 73, no. 4, pp. 213–227, 1998.
[52] P. Baranowski, J. Lipecki, W. Mazurek, and R. T. Walczak,
“Detection of watercore in ‘Gloster’apples using thermogra-
phy,”Postharvest Biology and Technology, vol. 47, no. 3,
pp. 358–366, 2008.
[53] P. Baranowski and W. Mazurek, “Detection of physiological
disorders and mechanical defects in apples using thermogra-
phy,”International Agrophysics, vol. 23, no. 1, pp. 9–17, 2009.
[54] T. Defraeye, V. Lehmann, D. Gross et al., “Application of MRI
for tissue characterisation of ‘Braeburn’apple,”Postharvest
Biology and Technology, vol. 75, pp. 96–105, 2013.
[55] K. Chigwaya, A. D. Plessis, D. W. Viljoen, I. J. Crouch, and
E. M. Crouch, “Use of X-ray computed tomography and 3D
image analysis to characterize internal browning in ‘Fuji’
apples after exposure to CO
2
stress,”Scientia Horticulturae,
vol. 277, Article ID 109840, 2021.
[56] D. E. Schut, R. M. Wood, A. K. Trull et al., “Joint 2D to 3D
image registration workflow for comparing multiple slice pho-
tographs and CT scans of apple fruit with internal disorders,”
Postharvest Biology and Technology, vol. 211, Article ID
112814, 2024.
11Journal of Food Processing and Preservation

[57] A. Tempelaere, L. Van Doorselaer, J. He, P. Verboven, and
B. M. Nicolai, “BraeNet: internal disorder detection in ‘Brae-
burn’apple using X-ray imaging data,”Food Control, vol. 155,
Article ID 110092, 2024.
[58] E. Herremans, P. Verboven, E. Bongaers et al., “Characterisa-
tion of ‘Braeburn’browning disorder by means of X-ray
micro-CT,”Postharvest Biology and Technology, vol. 75,
pp. 114–124, 2013.
[59] W. Chayaprasert and R. Stroshine, “Rapid sensing of internal
browning in whole apples using a low-cost, low-field proton
magnetic resonance sensor,”Postharvest Biology and Technol-
ogy, vol. 36, no. 3, pp. 291–301, 2005.
[60] N. Hernández-Sánchez, B. P. Hills, P. Barreiro, and
N. Marigheto, “An NMR study on internal browning in pears,”
Postharvest Biology and Technology, vol. 44, no. 3, pp. 260–
270, 2007.
[61] T. Vandendriessche, H. Schäfer, B. E. Verlinden, E. Humpfer,
M. L. Hertog, and B. M. Nicolaï, “High-throughput NMR
based metabolic profiling of Braeburn apple in relation to
internal browning,”Postharvest Biology and Technology,
vol. 80, pp. 18–24, 2013.
[62] Z. Zhang, Y. Pu, Z. Wei et al., “Combination of interactance
and transmittance modes of Vis/NIR spectroscopy improved
the performance of PLS-DA model for moldy apple core,”
Infrared Physics and Technology, vol. 126, Article ID 104366,
2022.
[63] V. A. Mcglone, P. J. Martinsen, C. J. Clark, and R. B. Jordan,
“On-line detection of Brownheart in Braeburn apples using
near infrared transmission measurements,”Postharvest Biol-
ogy and Technology, vol. 37, no. 2, pp. 142–151, 2005.
[64] J. H. Wang, X. D. Sun, L. Pan, Q. Sun, and D. H. Han, “Dis-
crimination of brownheart and watercore of apples based on
energy spectrum of visible/near infrared transmittance,”Spec-
troscopy and Spectral Analysis, vol. 28, no. 9, pp. 2098–2102,
2008.
[65] J. Sun, R. Künnemeyer, A. Mcglone, and N. Tomer, “Investiga-
tions of optical geometry and sample positioning in NIRS
transmittance for detecting vascular browning in apples,”
Computers and Electronics in Agriculture, vol. 155, pp. 32–40,
2018.
[66] Z. Guo, M. Wang, A. A. Agyekum et al., “Quantitative detec-
tion of apple watercore and soluble solids content by near
infrared transmittance spectroscopy,”Journal of Food Engi-
neering, vol. 279, Article ID 109955, 2020.
[67] Y. Zhang, Z. Wang, X. Tian, X. Yang, Z. Cai, and J. Li, “Online
analysis of watercore apples by considering different speeds
and orientations based on Vis/NIR full-transmittance spec-
troscopy,”Infrared Physics and Technology, vol. 122, Article
ID 104090, 2022.
[68] J. Li, Y. Zhang, Q. Zhang, D. Duan, and L. Chen, “Establish-
ment of a multi-position general model for evaluation of
watercore and soluble solid content in ‘Fuji’apples using on-
line full-transmittance visible and near infrared spectroscopy,”
Journal of Food Composition and Analysis, vol. 117, Article ID
105150, 2023.
[69] Z. Guo, Y. Zou, C. Sun et al., “Nondestructive determination of
edible quality and watercore degree of apples by portable Vis/
NIR transmittance system combined with CARS-CNN,”Jour-
nal of Food Measurement and Characterization, vol. 18, no. 6,
pp. 4058–4073, 2024.
[70] M. R. Mogollon, A. F. Jara, C. Contreras, and J. P. Zoffoli,
“Quantitative and qualitative VIS-NIR models for early deter-
mination of internal browning in ‘Cripps Pink’apples during
cold storage,”Postharvest Biology and Technology, vol. 161,
Article ID 111060, 2020.
[71] S. Tian, M. Zhang, B. Li et al., “Measurement orientation com-
pensation and comparison of transmission spectroscopy for
online detection of moldy apple core,”Infrared Physics &
Technology, vol. 111, Article ID 103510, 2020.
[72] Z. Hu, Y. Pu, W. Wu, L. Pan, Y. Yang, and J. Zhao, “Online
detection of moldy apple core based on diameter and SSC fea-
tures,”Food Control, vol. 168, Article ID 110879, 2025.
[73] X. Tian, Q. Wang, W. Huang, S. Fan, and J. Li, “Online detec-
tion of apples with moldy core using the Vis/NIR full-
transmittance spectra,”Postharvest Biology and Technology,
vol. 168, Article ID 111269, 2020.
[74] K. Zhang, H. Jiang, H. Zhang et al., “Online detection and clas-
sification of moldy core apples by Vis-NIR transmittance spec-
troscopy,”Agriculture, vol. 12, no. 4, p. 489, 2022.
[75] W. Guo, W. Li, B. Yang, Z. Zhu, D. Liu, and X. Zhu, “Anovel
noninvasive and cost-effective handheld detector on soluble
solids content of fruits,”Journal of Food Engineering, vol. 257,
pp. 1–9, 2019.
[76] A. Pissard, E. J. N. Marques, P. Dardenne et al., “Evaluation
of a handheld ultra-compact NIR spectrometer for rapid
and non-destructive determination of apple fruit quality,”
Postharvest Biology and Technology, vol. 172, Article ID
111375, 2021.
[77] Z. Guo, Y. Zhang, J. Wang et al., “Detection model transfer of
apple soluble solids content based on NIR spectroscopy and
deep learning,”Computers and Electronics in Agriculture,
vol. 212, Article ID 108127, 2023.
[78] H. Chang, J. Yin, H. Tian, J. Yan, and H. Xu, “Evaluation of the
optical layout and sample size on online detection of apple
watercore and SSC using Vis/NIR spectroscopy,”Journal of
Food Composition and Analysis, vol. 123, Article ID 105528,
2023.
[79] Y. Yang, C. Zhao, W. Huang et al., “Optimization and compen-
sation of models on tomato soluble solids content assessment
with online Vis/NIRS diffuse transmission system,”Infrared
Physics and Technology, vol. 121, Article ID 104050, 2022.
[80] Y. Guo, Y. Ni, and S. Kokot, “Evaluation of chemical compo-
nents and properties of the jujube fruit using near infrared spec-
troscopy and chemometrics,”Spectrochimica Acta Part A:
Molecular and Biomolecular Spectroscopy,vol.153,pp.79–86,
2016.
[81] X. Tian, Q. Wang, J. Li, F. Peng, and W. Huang, “Non-destruc-
tive prediction of soluble solids content of pear based on fruit
surface feature classification and multivariate regression anal-
ysis,”Infrared Physics and Technology, vol. 92, pp. 336–344,
2018.
[82] N. Zhao, Z. Wu, Y. Cheng, X. Shi, and Y. Qiao, “MDL and
RMSEP assessment of spectral pretreatments by adding different
noises in calibration/validation datasets,”Spectrochimica Acta
Part A: Molecular and Biomolecular Spectroscopy,vol.163,
pp. 20–27, 2016.
[83] C. Xiong, C. Liu, W. Liu et al., “Noninvasive discrimination
and textural properties of E-beam irradiated shrimp,”Journal
of Food Engineering, vol. 175, pp. 85–92, 2016.
[84] K. Yu, Y. Zhao, X. Li, Y. Shao, F. Zhu, and Y. He, “Identifica-
tion of crack features in fresh jujube using Vis/NIR hyperspec-
tral imaging combined with image processing,”Computers
and Electronics in Agriculture, vol. 103, pp. 1–10, 2014.
12 Journal of Food Processing and Preservation

[85] J.-H. Cheng and D.-W. Sun, “Partial least squares regression
(PLSR) applied to NIR and HSI spectral data modeling to pre-
dict chemical properties of fish muscle,”Food Engineering
Reviews, vol. 9, no. 1, pp. 36–49, 2016.
[86] D. Hatoum, M. L. Hertog, A. H. Geeraerd, and B. M. Nicolai,
“Effect of browning related pre- and postharvest factors on
the ‘Braeburn’apple metabolome during CA storage,”Post-
harvest Biology and Technology, vol. 111, pp. 106–116, 2016.
[87] J. J. Gonzalez, R. C. Valle, S. Bobroff, W. V. Biasi, E. J.
Mitcham, and M. J. Mccarthy, “Detection and monitoring of
internal browning development in ‘Fuji’apples using MRI,”
Postharvest Biology and Technology, vol. 22, no. 2, pp. 179–
188, 2001.
13Journal of Food Processing and Preservation