Académique Documents
Professionnel Documents
Culture Documents
Food Chemistry
journal homepage: www.elsevier.com/locate/foodchem
a r t i c l e i n f o a b s t r a c t
Article history: Evaluation of water status in Jonathan apples was performed for 20 days. Loss moisture content (LMC)
Available online 6 June 2013 was carried out through slow drying of wholes apples and the moisture content (MC) was carried out
through oven drying and lyophilisation for apple samples (chunks, crushed and juice).
Keywords: We approached a non-destructive method to evaluate LMC and MC of apples using image processing
Water status and multilayer neural networks (NN) predictor. We proposed a new simple algorithm that selects the tex-
Moisture content ture descriptors based on initial set heuristically chosen. Both structure and weights of NN are optimised
Jonathan apples
by a genetic algorithm with variable length genotype that led to a high precision of the predictive model
Imagistic analysis
Neural network
(R2 = 0.9534).
In our opinion, the developing of this non-destructive method for the assessment of LMC and MC (and
of other chemical parameters) seems to be very promising in online inspection of food quality.
Ó 2013 Elsevier Ltd. All rights reserved.
0308-8146/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.foodchem.2013.05.131
L.C. Trincă et al. / Food Chemistry 144 (2014) 80–86 81
2. Material and methods 5 g of fresh apple sample (chunks, crushed or juice) were sub-
jected to freeze drying at temperature of 50 to 60 °C, and pres-
Apples are very popular fruits in Europe. There are many variety sure ranging from 0.02 to 0.03 mbar until constant mass. Water
of apple fruits having different colours (green, red, yellow, or content was determined using the Eq. (2).
mixed red-yellow) and different changes in aspect and structure
during storage. 2.2. Methods for imagistic analysis
There are many approaches that deal with the subject of drying
fruits, most of them being focused on pieces of fruits (Fernández, Non-destructive methods are usually based on image process-
Castillero, & Aguilera, 2005; Seiiedlou, Ghasemzadeh, Hamdami, ing techniques (Gonzalez & Woods, 2008; Petrou & Sevilla, 2006;
Talati, & Moghaddam, 2010). Our approach is focused on the mod- Yam & Papadakis, 2004). The images are acquired under special
ification of the moisture content in apples during storage at ordin- conditions (illumination, angle between axis of camera lens and
ary temperature. the lighting source – 45°, in order to capture the diffused reflection,
At early November, 100 healthy-looking Jonathan apples, that colour temperature of the fluorescent lamp and so on.) in accor-
weren’t exposed to preservation treatments, were selected ran- dance with the image’s acquisition standards (Francis, 2005)
domly from the same source, with roughly the same shape and Analysis of texture can offer useful information about quality of
weight (91–103 g), as being the gross sample of analysis. Chemical fruits (Gonzalez & Woods, 2008). At apples, the wrinkles usually
and imagistic tests have been conducted every 5 days: lots of 15 denote long time storage: more wrinkles involve a larger storage
apples were investigated in every testing day for MC evaluation. period and a higher loss of the moisture content. Also, it looks like
Also, 15 other apples were analysed all experimental period for wrinkle’s texture parameters are related with the variation of LMC
LMC determination. All apples were stored during experimental (Vesali, Gharibkhani, & Komarizadeh, 2011)
period at 20 °C room temperature and 70% relative humidity. Fourier descriptors for 2-D digital images seem to be useful for
A balance type RADWAG AS 220/C/2 with accuracy of 104 g characterization of LMC. However, usually way to use Fourier
was used for weighing, a SLW 115 ECO drying oven was used for descriptors are related to shape shrinking due to drying process.
drying apple samples at temperature of 90 °C and a Freeze Dryer For rectangular pieces of apples, the change in shape due to drying
ALPHA type 1–4 LD was used for lyophilisation. process is significant. The evolution of shape in time can be mea-
For image capturing an AF-S DX Zoom-Nikon ED 18–70 mm, f/ sured if the shape is coded in a vector form by Fourier descriptors
3.5–4.5 GIF was used. (Fernández et al., 2005). In our case, these descriptors doesn’t offer
useful information related to LMC due faintly the fact that Jonathan
2.1. Methods for chemical analysis apple variety provides in time a very small change in their shape.
The experiment achievement until the moment that the shape of
Analyses were performed for a period of 20 days. For LMC eval- apple significantly changed was not of interest to us, because in
uation each entire apple was weighed and photographed. this case, the apple cannot be accepted by consumers.
Also, MC analysis had involved the weighing and photographing Unlike the data presented in the literature (Fernández et al.,
of each lot of 15 apples, in the first step. For MC evaluation, in the 2005; Seiiedlou et al., 2010) for which investigations have been
second step, chemical analyses were performed on the samples conducted on apple’s cuticle, our experiments focused on the im-
immediately obtained by processing the apple as chunks, crushed age acquired for the entire apple.
and juice. The apple has been peeled and then has been shredded Tests have been conducted every five days: lots of 15 apples
with a special device to get apple chunks (0.4 2 cm). Also, peeled were investigated in every testing day for MC evaluation. Also,
apple was crushed in order to get crushed apple or pressed to get 15 other apples were analysed all experimental period for LMC
apple juice. determination. For LMC evaluation each entire apple was photo-
graphed. Also, MC analysis had involved the photographing of each
lot of 15 apples, in the first step.
2.1.1. Loss moisture content determination
For image capturing an AF-S DX Zoom-Nikon ED 18–70 mm, f/
At every 5 days, apples were weighed regularly. By weighing
3.5–4.5 GIF was used. The pictures were saved in RGB colour space
the difference was attributed to the loss of moisture content
with 1936 1926 pixels size. In order to ensure the same calibra-
(LMC, %), in accordance with the Eq. (1):
tion of dimension in all pictures, a box was designed
LMC; % ¼ 100 ðmi mf Þ=mi ð1Þ (16 16 32 cm) to provide a fixed distance between photo cam-
era and apples.
where mi = initial mass of sample (g), mf = final mass of sample (g).
The photos were processed and specific texture parameters
were determined (see discussion at Section 3.2.) in order to build
2.1.2. Oven drying NN predictive models for LMC and MC of apples.
5 g of fresh sample (each apple was processed to get chunks,
crushed or juice) were dried at 90 °C until a constant mass 2.3. Predictive methods
6102 g was reached. Moisture content (MC, %) was determined
in accordance with the Eq. (2): 2.3.1. Neural networks predictive methods
MC; % ¼ 100 ðm m1 Þ=m2 ð2Þ Neural networks (NN) proven to be universal approximators
(Haykin, 1999). In 1989, Cybenko demonstrated, using a Kolmogo-
where m = mass of the sample before drying (g), m1 = mass of the rov’s older result, that multilayer feed-forward network with a sin-
sample after drying (g), m2 = mass of the analysed sample (g). gle hidden layer, which contains finite number of hidden neurons,
is a universal approximator (Cybenko, 1989). The demonstration
2.1.3. Freeze drying (lyophilisation) has been made for sigmoid activation function. In the most com-
Lyophilisation is dehydration process at cold: water freezes fas- mon sense, the universal approximator can approximate any
ter than other ingredients and is removed in the form of ice with- non-linear function with a desired precision, if its architecture is
out changing the food structure. large enough. We will use a multilayer neural network (multilayer
Lyophilisation was carried out in a type 1–4 LD plus, Freeze- perceptron, MLP) as predictor for LMC and MC. The structure and
Dryer ALPHA. weights of MLP are simultaneous adapted using a genetic
82 L.C. Trincă et al. / Food Chemistry 144 (2014) 80–86
6.5 Apple1
Apple2
6
Apple3
5.5
Apple4
5 Apple5
LMC, (%)
4.5 Apple6
Apple7
4
Apple8
3.5 Apple9
3 Apple10
2.5 Apple11
Apple12
2
Apple13
1.5 Apple14
1 Apple15
Day 5 Day 10 Day 15 Day 20 Average
Period of time, (days)
Fig. 3. Dynamics of loss moisture content (LMC) for the set of 15 apples.
Table 1
Loss moisture content (LMC,%) and moisture content (MC,%) of apples samples during the experiment.
Jonathan apple’s weight loss explains the LMC increasing, 1936 1926 pixels size (Fig. 5). From each apple four equal rectan-
although a very significant change doesn’t appeared in their gular areas that not include stem area (pedicel area) were denoted,
dimension. so, will result finally 90 4 = 360 samples.
During the experiment, images of 90 wholes apple were cap- The apples are approximated by spheres and the radius of this
tured and the pictures were saved in RGB colour space with sphere is used as form descriptor (FD) for prediction of LMC. The
84 L.C. Trincă et al. / Food Chemistry 144 (2014) 80–86
82 X
L
m¼ zi pðzi Þ ð3Þ
81 i¼1
1
MC (%)
80 S¼1 ð4Þ
1 þ r2
79
X
L
78
l3 ¼ ðzi IÞ3 pðzi Þ ð5Þ
i¼1
77 X
L
0 5 10 15 20 U¼ p2 ðzi Þ ð6Þ
Period of time (days) i¼1
Chunks Crushed X
L
Fig. 6. The G image of apple, the contour of apple (cyan) and the approximation by
a circle (blue). Fig. 7. The multilayer perceptron used for NN model predictor.
L.C. Trincă et al. / Food Chemistry 144 (2014) 80–86 85
Table 2
Performance of NN using inputs V = {m, S, l3, U, e, FD}, R2 values.
Model Method
Neural network Slow drying Oven drying Lyophilisation
LMC, % (whole apple) MC, % (chunks apple) MC, % (crushed apple) MC, % (juice apple) MC, % (chunks apple)
Learning stage 0.8941 0.9915 0.9915 0.9915 0.9915
Testing stage 0.6381 0.8335 0.8335 0.8335 0.8335
Table 3
Performance of NN using inputs V = {m, S, 3, U, e, FD, Hk}, R2 values.
Table 4
Performance of NN using inputs V = {m, S, 3, U, e, FD, H8}, R2 values.
Model Method
Neural network Slow drying Oven drying Lyophilisation
LMC, % (whole apple) MC, % (chunks apple) MC, % (crushed apple) MC, % (juice apple) MC,% (chunks apple)
Learning stage 0.9989 0.9989 0.9989 0.9989 0.9989
Testing stage 0.9534 0.9534 0.9534 0.9534 0.9534
The predicted outcome of NN can be one of the variables: LMC the algorithm. If no improvement is made, take into account
(whole apples) or MC (chunks/crushed/ juice/lyophilised apple the list Vinput Hp and stop the algorithm. If the computational
samples). effort is acceptable, go to Step 3.
Step 6. Clear Hp from list Lh.
3.3.2. Genetic algorithm with variable length genotype
The results for learning stage and testing case using V = {m, S, The algorithm GAVLG runs separately for each population of
l3, U, e, FD} are shown in Table 2. chromosomes. We have three populations of chromosomes and
We propose a novel algorithm to improve the results. We use we start with 50 individuals, with lengths between 4 and 10
the 13th texture descriptors proposed by Haralick, Shanmugam, weights. These values have been chosen by empirical assertions,
and Dinstein (1973), denoted by Hk, k = 1 .... 13. We eliminated H9 based on programmer’s experience. After 40 generations, an opti-
(entropy) from the list because this term is already taken in V. mal architecture has been found: 15 nodes in the first hidden layer
Main steps of GAVLG algorithm for prediction of LMC/MC are: and 10 nodes in the second hidden layer.
We can see from Table 3 that if we append the H8 descriptor,
Step 1. Set the basic list Vinput = V and additional list Hk, k2Lh, the precision of fitting increase from 0.6381 (that is an unaccept-
Lh = {1, ... ,8,10, ... 13}. able value for a good model) to 0.9534, a good value for the model.
Step 2. Run learn and test stage for NN using Vinput list. Obtain R2 H8 refers to the sum of entropy:
for LMC.
Step 3. For all, k2Lh run learn and test stage for NN using X
2L
Vinput = V[Hk. H8 ¼ pxþy ðzi Þ log pxþy ðzi Þ ð8Þ
Step 4. The descriptors Hk that produced a greater value than in i¼2
In accordance with step 5 of GAVLG algorithm the computa- Francis, F. J. (2005). Colorimetric Property of Foods. In M. A. Rao, S. S. H. Rizvi, & A. K.
Datta (Eds.), Engineering Property of Foods (pp. 703–732). Boca-Raton, Florida:
tional effort in this case is substantially increased, so, the optimal
CRC Press, Taylor & Francis Group.
Vinput can be considered V = {m, S, l3, U, e, FD, H8}. Goñi, O., Muñoz, M., Ruiz-Cabello, J., Escribano, M. I., & Merodio, C. (2007). Changes
in water status of cherimoya fruit during ripening. Postharvest Biology and
Technology, 45, 147–150.
Gonzalez, R. C., & Woods, R. E. (2008). Digital image processing (3th ed.). Upper
4. Conclusions Saddle River, New Jersey: Prentice Hall.
Haralick, R. M., Shanmugam, K., & Dinstein, I. (1973). Textural features for image
Evaluation of water status in Jonathan apples was performed for classification. IEEE Transaction on Systems, Man and Cybernetics, SMC-3(6),
610–621.
twenty days. Loss moisture content (LMC) was carried out through Haykin, S. (1999). Neural networks: A comprehensive foundation (2th ed.). Upper
slow drying of wholes apples and the moisture content (MC) was Saddle River, New Jersey: Prentice Hall.
carried out through oven drying and lyophilisation for apple sam- Hills, B. P., & Remigereau, B. (1997). NMR studies of changes in subcellular water
compartmentation in parenchyma apple tissue during drying and freezing.
ples (chunks, crushed and juice).
International Journal of Food Science and Technology, 32, 51–61.
We approached a non-destructive method to evaluate LMC and Isengard, H.-D., Merkh, G., Schreib, K., Labitzke, I., & Dubois, C. (2010). The influence
MC of apples using image processing and multilayer neural net- of the reference method on the results of the secondary method via calibration.
Food Chemistry, 122(2), 429–435.
works (NN) predictor.
Jan, I., & Rab, A. (2012). Influence of storage duration on physico-chemical changes
We proposed a new simple algorithm that selects the texture of fruits of apples cultivars. The Journal of Animal and Plant Sciences, 22(3),
descriptors based on initial set heuristically chosen. Both structure 708–714.
and weights of NN are optimised by a genetic algorithm with var- The MathWorks, Inc. MATLAB program. URL <http://www.mathworks.com/>.
Accessed 5.10.12.
iable length genotype that led to a high precision of the predictive Menges, H. O., & Ertekin, C. (2006). Mathematical modeling of thin layer drying of
model (R2 = 0.9534). Golden apples. Journal of Food Engineering, 77(1), 119–125.
In our opinion, the developing of this non-destructive method Munir-ul, M., Chowdhury, M., & Yun Li (1996). Messy genetic algorithm based new
learning method for structurally optimised neurofuzzy controllers. Proceedings
for the assessment of LMC and MC (and of other chemical param- of The IEEE International Conference on Industrial Technology (ICIT ‘96). 274–278.
eters) seems to be very promising in online inspection of food Nguyen, T. A., Dresselaers, T., & Verboven, P. (2006). Finite element modelling and
quality. MRI validation of 3D transient water profiles in pears during postharvest
storage. Journal of the Science of Food and Agriculture, 86, 745–756.
Okos, M. R., Osvaldo, C., Campanella, O., Narsimhan, G., Singh, R. K., & Weitnauer, A.
(2007). Food dehydration. In D. R. Heldman & D. B. Lund (Eds.), Handbook of food
References engineering (pp. 601–743). Boca Raton: CRC Press, Taylor and Francis Group.
Petrou, M., & Sevilla, P. G. (2006). Image processing: Dealing with texture. West
Bradley, R. L. Jr., (2010). Moisture and total solids analysis. In S. S. Nielsen (Ed.), Food Sussex: John Wiley & Sons, Ltd..
analysis (pp. 85–104). New York: Springer Science, Business Media, LLC. Rahman, M. S. (2006). State diagram of foods: Its potential use in food processing
Cybenko, G. (1989). Approximations by superpositions of a sigmoidal function. and product stability. Trends in Food Science and Technology, 17, 129–141.
Mathematics of Control, Signals, and Systems, 2(4), 303–314. Schmidt, S. J. (2007). Water mobility in foods. In G. V. Barbosa-Cánovas, A. J.
Derossi, A., Severini, C., & Cassi, D. (2011). Mass transfer mechanisms during Fontana, Jr., S. J. Schmidt, & T. P. Labuza (Eds.), Water activity in foods:
dehydration of vegetable food: Traditional and innovative approaches. In M. El- Fundamentals and applications (pp. 47–108). Ames, Iowa: IFT Press & Blackwell
Amin (Ed.), Advanced topics in mass transfer (pp. 305–354). Rijeko: InTech – Publishing.
Open Access Publisher. Seiiedlou, S., Ghasemzadeh, H. R., Hamdami, N., Talati, F., & Moghaddam, M. (2010).
Devore L.J. Berk N.K. Modern mathematical statistics with applications 2007, Convective drying of apple: Mathematical modeling and determination of some
Belmont: Duxbury. quality parameters. International Journal of Agriculture and Biology, 12(2),
El-Sayd, N. I., & Makawy, M. M. (2010). Comparison of methods for determination of 171–178.
moisture in food. Research Journal of Agriculture and Biological Sciences, 6(6), Veraverbeke, E., Lammertyn, J., Saevels, S., & Nicolaï, B. (2001). Changes in chemical
906–911. wax composition of three different apple (Malus domestica Borkh.) cultivars
Erturk, U., Akbuclak, B., & Ozer, M. H. (2003). Quality changes of some apple during storage. Postharvest Biology and Technology, 23(3), 197–208.
cultivars stored in normal atmosphere for long period. Acta Horticulturae, 599, Veraverbeke, E., Verboven, P., Van Oostveldt, P., & Nicolaï, B. (2003). Prediction of
665–672. moisture loss across the cuticle of apple (Malus sylvestris subsp mitis (Wallr.)
Fernández, L., Castillero, C., & Aguilera, J. M. (2005). An application of image analysis during storage: Part 2. Model simulations and practical applications. Postharvest
to dehydration of apple discs. Journal of Food Engineering, 67(1–2), 185–193. Biology and Technology, 30(1), 89–97.
Fontana, A. J. Jr., (2007). Measurement of water activity, moisture sorption Vesali, F., Gharibkhani, M., & Komarizadeh, M. H. (2011). An approach to estimate
isotherms and moisture content of foods. In G. V. Barbosa-Cánovas, A. J. moisture content of apple with image processing method. Australian Journal of
Fontana, Jr., S. J. Schmidt, & T. P. Labuza (Eds.), Water activity in foods: Crop Science, 5(2), 111–115.
Fundamentals and applications (pp. 155–173). Ames, Iowa: IFT Press, Blackwell Yam, K. L., & Papadakis, S. E. (2004). A simple digital imaging method for measuring
Publishing. and analyzing color of food surfaces. Journal of Food Engineering, 61(1), 137–142.