Available via license: CC BY 4.0
Content may be subject to copyright.
Citation: Gordillo, M.C.L.;
Madueño-Luna, A.; Luna, J.M.M.;
Ramírez-Juidías, E. Use of Artificial
Vision during the Lye Treatment of
Sevillian-Style Green Olives to
Determine the Optimal Time for
Terminating the Cooking Process.
Foods 2023,12, 2815. https://
doi.org/10.3390/foods12142815
Academic Editor: Billy Hammond
Received: 6 June 2023
Revised: 1 July 2023
Accepted: 22 July 2023
Published: 24 July 2023
Copyright: © 2023 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
foods
Article
Use of Artificial Vision during the Lye Treatment of
Sevillian-Style Green Olives to Determine the Optimal Time
for Terminating the Cooking Process
Miguel Calixto López Gordillo 1, Antonio Madueño-Luna 2, *, JoséMiguel Madueño Luna 1
and Emilio Ramírez-Juidías 1
1Graphics Engineering Department, University of Seville, 41013 Seville, Spain; mlopez@us.es (M.C.L.G.);
jmadueno@us.es (J.M.M.L.); erjuidias@us.es (E.R.-J.)
2Aeroespace Engineering and Fluid Mechanical Department, University of Seville, 41013 Seville, Spain
*Correspondence: amadueno@us.es
Abstract:
This study focuses on characterizing the temporal evolution of the surface affected by
industrial treatment with NaOH within the processing tanks during the lye treatment stage of
Manzanilla table olives. The lye treatment process is affected by multiple variables, such as ambient
temperature, the initial temperature of the olives before lye treatment, the temperature of the NaOH
solution, the concentration of the solution, the variety of olives, and their size, which are determinants
of the speed of the lye treatment process. Traditionally, an expert, relaying on their subjective
judgement, manages the cooking process empirically, leading to variability in the termination timing
of the cook. In this study, we introduce a system that, by using an artificial vision system, allows us
to know in a deterministic way the percentage of lye treatment achieved at each moment along the
cooking process; furthermore, with an interpolator that accumulates values during the lye treatment,
it is possible to anticipate the completion of the cooking by indicating the moment when two-thirds,
three-fourths, or some other value of the interior surface will be reached with an error of less than
10% relative to the optimal moment. Knowing this moment is crucial for proper processing, as it will
affect subsequent stages of the manufacturing process and the quality of the final product.
Keywords: table olive; sodium hydroxide; artificial vision; data interpolation
1. Introduction
The olive tree (Olea europaea) is a small evergreen tree that is long-lived and can reach
up to 15 m in height. It has a wide crown and a thick, twisted trunk, and its fruit is the
olive [
1
]. The olive has been used both for oil extraction and as a food component within
the Mediterranean diet [2].
The global production of table olives varies from year to year and depends on several
factors, such as climate, diseases, and economic conditions. According to the International
Olive Council [
3
], Spain produced about 2.7 million tons in the 2020/2021 campaign,
which represents 20.5% of the world’s table olives, with a volume 19.3% higher than the
previous year.
The industrial-scale preparation of the first table olives in Spain, “Spanish or Sevillian-
style green olives”, began in the late 19th century in the province of Seville, in the towns of
Dos Hermanas, Alcaláde Guadaíra, Morón, and Arahal, among others.
These Sevillian-style green table olives [4] are made from olive varieties harvested in
the stage before ripening, with the desired consistency and size; the heat transfer during
its cooking has been studied [
4
] due to the costs of the energy involved in the treatment.
These costs are currently rising; thus, optimizing them is in the interest of this industry.
The industrial process consists of two fundamental stages: lye treatment (to remove the
bitterness due to oleuropein) [
5
,
6
] and lactic fermentation in brine. The lye treatment
Foods 2023,12, 2815. https://doi.org/10.3390/foods12142815 https://www.mdpi.com/journal/foods
Foods 2023,12, 2815 2 of 13
stage involves treating the fruits with an aqueous solution of sodium hydroxide (NaOH)
at 2–4% (w/v) for about 6–12 h, depending on the variety, the ripeness of the fruits, and
the temperature of the solution. Once the diffusion of the alkali has reached between
two-thirds and three-fourths of the pulp, the treatment is stopped, and the NaOH solution
and excess alkali are removed by one or several static water washes. Finally, the olives are
immersed in a brine solution with a concentration of 10–12% (w/v) sodium chloride, where
lactic fermentation takes place to prevent the growth of undesirable microorganisms and
promote the growth of lactobacilli [7].
The tanks used for the lye treatment of the olives and containing NaOH and NaCl
are fiberglass and polyester containers, conical at the ends and cylindrical in the central
part, with a capacity of 16,000 L (approximately 10,000 kg of olives and 6000 L of alkali
capacity, in the case of lye treatment tanks). The tanks are usually located inside the
industry, and the lye treatment processes take place preferably at night so that the olives
arrive with a lower temperature and the lye solution does not need to be cooled so much to
achieve a predetermined equilibrium temperature for cooking. This represents a source
of savings for the factory in refrigeration as it does not have to lower the temperature
of the NaOH solution so much. The concentration of the alkali used for lye treatment is,
on average, 3.0%
±
0.2% (w/v) in the tests described in this work. In current practice, as
the preparation of Sevillian-style olives has extended to other varieties and, therefore, to
other areas different from Seville, the range of lye concentrations used has greatly varied to
maintain approximately the same number of hours for cooking.
The nutritional and sensory quality of the product depend on the cultivation condi-
tions, mainly regulated irrigation deficits [
8
], pre-lye treatments (especially during transport
in low-concentration alkaline solutions) [
9
], and the conditions of the NaOH treatment
itself [10–17].
A cooking process where optimal times are not respected can lead to significant
defects [
7
]; for example, in the Manzanilla variety, when the lye penetrates up to the middle
of the pulp, the fruits remain bitter for a long time after being in brine, with the pulp near
the bone having a more or less brown color. When penetration is equal to or greater than
two-thirds, these issues do not occur. This phenomenon of abnormal colors, especially with
olives from the Manzanilla variety from irrigated sources, is related to the greater resulting
heterogeneity in the degree of maturity of these fruits, making it more difficult to determine
the end point of cooking. In later stages such as packaging, these colors spread to the skin
of the fruit, resulting in a loss of organoleptic quality.
Another notable example is that, if the degree of penetration of the lye is excessive,
the olives break during pitting; on the contrary, if it is insufficient, the pit does not come
out clean and drags a large part of the pulp.
To ensure that all olives achieve appropriate penetration simultaneously [
1
], the
batches of this fruit designated for cooking must be as uniform as possible in size and
ripeness. Once the cooking process has been completed and the lye has been removed,
the olives are covered with water with the aim of removing the largest possible amount of
sodium hydroxide that coats the olives and that has also penetrated the pulp. Care must
be taken not to over-wash the olives to avoid loss of soluble compounds necessary for
subsequent fermentation.
The skin of the olives offers considerable resistance to the attack of the lye, with a
somewhat long initial time, depending on factors such as variety, maturity, lye concentra-
tion, and temperature. During this time, the lye only acts on the skin and does not penetrate
the pulp of the olive.
The washing process [
7
] is a complementary operation to the cooking process, and it
serves to remove the lye that remains adhered to the surface of the fruits and a part of the
one that penetrated inside the pulp. Another of its objectives is to partially eliminate the
bitterness produced by oleuropein by hydrolysis [18].
Once the water washing is completed, the olives are placed in a brine with concentra-
tions between 10% and 12% (w/v), where they are maintained during the fermentation and
Foods 2023,12, 2815 3 of 13
preservation stages. This fermentation process is typically conducted in buried containers,
unlike the cooking process which is usually carried out in aerial ones.
A long and vigorous wash removes all traces of bitterness and lye, resulting in low
pH values during subsequent fermentation, while eliminating sugars and other substances
that facilitate this fermentation. A short wash produces bitter olives and high pH values
that make the preservation of the olives difficult.
If these conditions are not respected [
1
], problems can occur during fermentation
such as (1) fissures on the exterior of the olives and interior cavities in the pulp due to the
formation of CO
2
-filled blisters under the skin, a phenomenon known as “alambrado” in
Spanish, (2) softening due to excessive development of microorganisms with pectinolytic
activity, (3) butyric fermentation by clostridia, (4) sedimentation during packaging caused
by propionic bacteria consuming lactic acid, and (5) a condition known as “zapatería” in
Spanish, which occurs during conservation and is caused by clostridia.
The maximum recommended duration for the cooking and washing process is 24 h [
1
],
although the number of washes is variable, and the current trend, considering the scarcity
of water and the pollution produced by these discharges, is to give a single wash of about
12–15 h.
In the scientific literature, it is possible to find specific examples of the use of artificial
vision with olives. For instance, the authors of [
19
] designed a system to classify olives
into four categories, while also analyzing the effectiveness of various algorithms [
20
]. The
fruit quality was evaluated and classified in [
21
,
22
] by establishing the external appearance
of the skin as a determining factor for fruit quality. Some authors pointed to the use of
infrared light for this purpose [
23
] or to estimate the ripening stage of olive lots [
24
]. Other
authors performed high-speed inspection of olives in DRR machines using artificial vision
and fast neural networks [
25
,
26
] or deep neural networks, such as Mask R-CNN [
27
] and
CNN algorithms, or using AI algorithms and RGB imaging [28].
This article describes the application of a new system [
29
] that deterministically
characterizes the lye treatment percentage through the use of artificial vision, as opposed
to relying on the subjective judgment of an expert “maestro cocedero” in Spanish, which
avoids all the problems described above. To do this, we use olive samples extracted from
the tank, to which phenolphthalein is applied to emphasize the contrast between areas
affected or not by NaOH. Subsequently, an image is generated and processed by artificial
vision to quantify the lye treatment percentage. The obtained pairs (time and percentage)
are used to feed the input of an interpolator. With the accumulated values during the
lye treatment process, it is possible to anticipate its completion, indicating the moment
when two-thirds, three-fourths, or another value of the surface affected by NaOH will be
reached, with an error less than 10% in the estimation of the optimal moment to end the
cooking process.
2. Materials and Method
2.1. Olive Variety Used
For the tests, olives from the Manzanilla de Sevilla variety [
1
,
30
–
35
] were used, char-
acterized by their nearly round shape and relatively small pit, presenting a high pulp/pit
ratio. It is one of the most productive and high-quality fruit varieties, which has facilitated
its international spread. It presents a problem of “peeling” when treated with lye imme-
diately after harvesting, which is why it is customary to let it rest for 2–3 days in small,
well-ventilated containers [
36
] or, alternatively, to pretreat it with diluted lye solutions
(0.5 (w/v)) for about 3–6 h.
2.2. Obtaining NaOH-Treated Olive Samples
The trials were carried out in a factory in Monturque (Córdoba, Spain), along with
another in Ferreira do Antalejo (Beja, Portugal), during the 2017 campaign, focusing
exclusively on the Manzanilla variety from Seville. The process of cooking olives with
NaOH (Brenntag Química S.A.U. Dos Hermanas, Seville, Spain), is carried out 24 h a day
Foods 2023,12, 2815 4 of 13
during the harvesting season. If performed during the night (as mentioned in Section 1), an
additional advantage in energy savings is obtained. During the lye treatment process, the
temperature in the tanks and the ambient temperature inside the lye treatment warehouse
were monitored. Traceability was followed, indicating the concentration of NaOH, the
temperature at which it was added to the tank, and the average size of the olives in the
tank. Through this monitoring, the entire lye treatment process could be analyzed, from
the beginning when adding the soda to the end when adding the washing water. In all
cases, fine Manzanilla variety olives were used with a size of 200–210 fruits/kg, along with
refrigerated NaOH (10–12
◦
C) at a concentration of (3–3.1% w/v). Olive samples were
extracted from the tanks every hour, and the following procedure was followed: (1) cutting
them axially flush with the pit, (2) applying a spray of phenolphthalein to obtain an intense
coloration in the area affected by NaOH, (3) allowing them to dry for 5 min, (4) placing
them in a scanner forming a matrix of four rows by eight columns (other arrangements
such as 4
×
6 can also be used depending on the method employed to obtain the image of
the olives; see Section 2.3), where the background is replaced by a black-colored surface,
ensuring identical lighting and image size conditions between samples, Figure 1.
Foods 2023, 12, 2815 4 of 13
2.2. Obtaining NaOH-Treated Olive Samples
The trials were carried out in a factory in Monturque (Córdoba, Spain), along with
another in Ferreira do Antalejo (Beja, Portugal), during the 2017 campaign, focusing ex-
clusively on the Manzanilla variety from Seville. The process of cooking olives with NaOH
(Brenntag Química S.A.U. Dos Hermanas, Seville, Spain), is carried out 24 h a day during
the harvesting season. If performed during the night (as mentioned in Section 1), an addi-
tional advantage in energy savings is obtained. During the lye treatment process, the tem-
perature in the tanks and the ambient temperature inside the lye treatment warehouse
were monitored. Traceability was followed, indicating the concentration of NaOH, the
temperature at which it was added to the tank, and the average size of the olives in the
tank. Through this monitoring, the entire lye treatment process could be analyzed, from
the beginning when adding the soda to the end when adding the washing water. In all
cases, fine Manzanilla variety olives were used with a size of 200–210 fruits/kg, along with
refrigerated NaOH (10–12 °C) at a concentration of (3–3.1% w/v). Olive samples were ex-
tracted from the tanks every hour, and the following procedure was followed: (1) cutting
them axially flush with the pit, (2) applying a spray of phenolphthalein to obtain an in-
tense coloration in the area affected by NaOH, (3) allowing them to dry for 5 min, (4)
placing them in a scanner forming a matrix of four rows by eight columns (other arrange-
ments such as 4 × 6 can also be used depending on the method employed to obtain the
image of the olives; see Section 2.3), where the background is replaced by a black-colored
surface, ensuring identical lighting and image size conditions between samples, Figure 1.
Figure 1. Sample of olives in 4 × 8 format colored with phenolphthalein on a dark background.
2.3. Image Capture System for Olives during the Lye Treatment Phase
The system consists of a black alveolar support designed by 3D printing and pre-
pared to accommodate the olives; the size of the alveoli depends on the caliber of the ol-
ives (Figure 2). The black color allows contrasting the background with the olives and thus
defining their contour precisely.
Figure 1. Sample of olives in 4 ×8 format colored with phenolphthalein on a dark background.
2.3. Image Capture System for Olives during the Lye Treatment Phase
The system consists of a black alveolar support designed by 3D printing and prepared
to accommodate the olives; the size of the alveoli depends on the caliber of the olives
(Figure 2). The black color allows contrasting the background with the olives and thus
defining their contour precisely.
Foods 2023, 12, 2815 5 of 13
Figure 2. Honeycomb support designed with 3D printing to hold up to 4 × 8 olives.
To maintain constant lighting, the support is placed inside a box whose ceiling con-
tains an LED lighting system (Figure 3) with the ability to adjust light intensity [37], using
a constant electrical intensity source [38,39].
Figure 3. Receptacle for maintaining constant lighting (with camera, smartphone [the one we used],
and scanner) and LED lighting system.
2.4. Artificial Vision Analysis System
For the image analysis, a procedure and software developed at the University of Se-
ville [29] were used, through which the green and red channels of the RGB image were
analyzed, as they contain information related to the penetration process of NaOH into the
pulp; the blue channel was used as an element to segment the background. The images
were acquired in two formats: (1) JPEG at 75 × 75 pixels per inch horizontally and verti-
cally, respectively, and with an image size of 1500 × 1125 pixels; (2) JPEG at 150 × 150 pixels
per inch horizontally and vertically, respectively, with an image size of 900 × 900 pixels,
proving to be a sufficient resolution for the artificial vision determination process of the
cooked surface with lye. Figure 4 shows the intermediate phases of the process for a 4 × 8
format image: segmentation of the olives from the background, identification of the part
attacked by NaOH, and determination of the cooked percentage.
Photo camera
Smartphone
Led
diodes
Olive tray
Scanner Constant intensity illuminated
supports adapted to digital
cameras, smartphones and
digital scanners
Figure 2. Honeycomb support designed with 3D printing to hold up to 4 ×8 olives.
Foods 2023,12, 2815 5 of 13
To maintain constant lighting, the support is placed inside a box whose ceiling contains
an LED lighting system (Figure 3) with the ability to adjust light intensity [
37
], using a
constant electrical intensity source [38,39].
Foods 2023, 12, 2815 5 of 13
Figure 2. Honeycomb support designed with 3D printing to hold up to 4 × 8 olives.
To maintain constant lighting, the support is placed inside a box whose ceiling con-
tains an LED lighting system (Figure 3) with the ability to adjust light intensity [37], using
a constant electrical intensity source [38,39].
Figure 3. Receptacle for maintaining constant lighting (with camera, smartphone [the one we used],
and scanner) and LED lighting system.
2.4. Artificial Vision Analysis System
For the image analysis, a procedure and software developed at the University of Se-
ville [29] were used, through which the green and red channels of the RGB image were
analyzed, as they contain information related to the penetration process of NaOH into the
pulp; the blue channel was used as an element to segment the background. The images
were acquired in two formats: (1) JPEG at 75 × 75 pixels per inch horizontally and verti-
cally, respectively, and with an image size of 1500 × 1125 pixels; (2) JPEG at 150 × 150 pixels
per inch horizontally and vertically, respectively, with an image size of 900 × 900 pixels,
proving to be a sufficient resolution for the artificial vision determination process of the
cooked surface with lye. Figure 4 shows the intermediate phases of the process for a 4 × 8
format image: segmentation of the olives from the background, identification of the part
attacked by NaOH, and determination of the cooked percentage.
Photo camera
Smartphone
Led
diodes
Olive tray
Scanner Constant intensity illuminated
supports adapted to digital
cameras, smartphones and
digital scanners
Figure 3.
Receptacle for maintaining constant lighting (with camera, smartphone [the one we used],
and scanner) and LED lighting system.
2.4. Artificial Vision Analysis System
For the image analysis, a procedure and software developed at the University of
Seville [
29
] were used, through which the green and red channels of the RGB image were
analyzed, as they contain information related to the penetration process of NaOH into the
pulp; the blue channel was used as an element to segment the background. The images
were acquired in two formats: (1) JPEG at 75
×
75 pixels per inch horizontally and vertically,
respectively, and with an image size of 1500
×
1125 pixels; (2) JPEG at 150
×
150 pixels
per inch horizontally and vertically, respectively, with an image size of 900
×
900 pixels,
proving to be a sufficient resolution for the artificial vision determination process of the
cooked surface with lye. Figure 4shows the intermediate phases of the process for a 4
×
8
format image: segmentation of the olives from the background, identification of the part
attacked by NaOH, and determination of the cooked percentage.
Foods 2023, 12, 2815 6 of 13
Figure 4. Image analysis process with artificial vision in 4 × 8 format.
2.5. Interpolator for Predicting the Optimal Moment to End the Lye Treatment Process
The evolution of the lye treatment curve, as seen in Figure 5, includes the following
phases: the beginning of the process with a slow speed and reduced slope, which coin-
cides with the filling of the tank with refrigerated lye [4] (in this phase, the olive skin is
permeated by the action of caustic soda); the central phase with high speed and increasing
temperature; the saturation phase due to lye consumption and the end of the alkaline pro-
cess with the addition of washing water. From this point on, it can be considered that the
penetration speed of NaOH inside the olive decreases considerably and stops completely
when the olives are poured into the fermentation brine.
To determine the optimal moment to end the lye treatment process, an interpolator
[40] is used, developed using the interp1 function in Matlab R2022b [41] (Figure 5). This
interpolator allows estimating, on the basis of a previously set lye treatment percentage,
the necessary lye treatment time in NaOH. It is important to note that the actual end time
of the lye treatment process is determined by experts according to their experience, which
may vary slightly from the theoretical prediction of the preset lye treatment percentage.
Figure 5. Example of how the lye treatment percentage evolves with a threshold at 84.3% when
proceeding to washing with water. This sigmoid curve has an initial phase of slow growth because
the olive skin has not yet been penetrated by the caustic soda.
The interpolation performed is based on the fact that we know two limits: first, at the
beginning of the process, the lye treatment percentage is 0%; second, at the end, if the
process is extended much longer than the optimal time (when the washing water is
ORIGINAL IMAGE CONTOURS LYE TREATMENT 76.2%
Figure 4. Image analysis process with artificial vision in 4 ×8 format.
2.5. Interpolator for Predicting the Optimal Moment to End the Lye Treatment Process
The evolution of the lye treatment curve, as seen in Figure 5, includes the following
phases: the beginning of the process with a slow speed and reduced slope, which coincides
Foods 2023,12, 2815 6 of 13
with the filling of the tank with refrigerated lye [
4
] (in this phase, the olive skin is permeated
by the action of caustic soda); the central phase with high speed and increasing temperature;
the saturation phase due to lye consumption and the end of the alkaline process with the
addition of washing water. From this point on, it can be considered that the penetration
speed of NaOH inside the olive decreases considerably and stops completely when the
olives are poured into the fermentation brine.
Foods 2023, 12, 2815 6 of 13
Figure 4. Image analysis process with artificial vision in 4 × 8 format.
2.5. Interpolator for Predicting the Optimal Moment to End the Lye Treatment Process
The evolution of the lye treatment curve, as seen in Figure 5, includes the following
phases: the beginning of the process with a slow speed and reduced slope, which coin-
cides with the filling of the tank with refrigerated lye [4] (in this phase, the olive skin is
permeated by the action of caustic soda); the central phase with high speed and increasing
temperature; the saturation phase due to lye consumption and the end of the alkaline pro-
cess with the addition of washing water. From this point on, it can be considered that the
penetration speed of NaOH inside the olive decreases considerably and stops completely
when the olives are poured into the fermentation brine.
To determine the optimal moment to end the lye treatment process, an interpolator
[40] is used, developed using the interp1 function in Matlab R2022b [41] (Figure 5). This
interpolator allows estimating, on the basis of a previously set lye treatment percentage,
the necessary lye treatment time in NaOH. It is important to note that the actual end time
of the lye treatment process is determined by experts according to their experience, which
may vary slightly from the theoretical prediction of the preset lye treatment percentage.
Figure 5. Example of how the lye treatment percentage evolves with a threshold at 84.3% when
proceeding to washing with water. This sigmoid curve has an initial phase of slow growth because
the olive skin has not yet been penetrated by the caustic soda.
The interpolation performed is based on the fact that we know two limits: first, at the
beginning of the process, the lye treatment percentage is 0%; second, at the end, if the
process is extended much longer than the optimal time (when the washing water is
ORIGINAL IMAGE CONTOURS LYE TREATMENT 76.2%
Figure 5.
Example of how the lye treatment percentage evolves with a threshold at 84.3% when
proceeding to washing with water. This sigmoid curve has an initial phase of slow growth because
the olive skin has not yet been penetrated by the caustic soda.
To determine the optimal moment to end the lye treatment process, an interpolator [
40
]
is used, developed using the interp1 function in Matlab R2022b [
41
] (Figure 5). This
interpolator allows estimating, on the basis of a previously set lye treatment percentage,
the necessary lye treatment time in NaOH. It is important to note that the actual end time
of the lye treatment process is determined by experts according to their experience, which
may vary slightly from the theoretical prediction of the preset lye treatment percentage.
The interpolation performed is based on the fact that we know two limits: first, at
the beginning of the process, the lye treatment percentage is 0%; second, at the end, if
the process is extended much longer than the optimal time (when the washing water is
added), it is 100% (for example, after 24 h). This allows us to dynamically correct the
interpolator’s prediction and get a good approximation of the completion time even from
the sixth sampling.
2.6. Cloud Computing System for Commercial-Level Analysis during Production
To manage the lye treatment process remotely and at a commercial level, a web
application (lye treatment) was developed using Matlab R2022b, Glassfish 5.0, Java JDK
1.7.0, and MySQL 5.7, capable of analyzing the process online, as shown in Figure 6a,b.
Foods 2023,12, 2815 7 of 13
Foods 2023, 12, 2815 7 of 13
added), it is 100% (for example, after 24 h). This allows us to dynamically correct the in-
terpolator’s prediction and get a good approximation of the completion time even from
the sixth sampling.
2.6. Cloud Computing System for Commercial-Level Analysis during Production
To manage the lye treatment process remotely and at a commercial level, a web ap-
plication (lye treatment) was developed using Matlab R2022b, Glassfish 5.0, Java JDK
1.7.0, and MySQL 5.7, capable of analyzing the process online, as shown in Figure 6a,b.
(a)
(b)
Figure 6. Web application for remote management of lye treatment. (a) Flowchart of the web ap-
plication operation. (b) Login screen of the web application.
3. Results
3.1. Laboratory Tests
Figure 7 shows one of the cases studied for the lye treatment (4 × 6 format). The total
duration of the process was 8 h, starting at 23:00 h with the addition of refrigerated NaOH
and ending at 7:30 h with the final wash with water. The results shown for each sample
were obtained with a 1 h interval, including the final washing of the olives with water
(sample number 8 in Figure 7).
As can be seen, the procedure followed for obtaining the images ensures a constant
background and lighting that allows a clear distinction between the olives and the back-
ground. The contrast provided by the phenolphthalein after the drying process allows for
a perfect contrast between the area affected by NaOH and the area that has not yet been
affected.
Figure 7. The 4 × 6 format images of olives at different moments of the lye treatment process with
NaOH; the last image corresponds to the phase of washing the olives in water.
Figure 6.
Web application for remote management of lye treatment. (
a
) Flowchart of the web
application operation. (b) Login screen of the web application.
3. Results
3.1. Laboratory Tests
Figure 7shows one of the cases studied for the lye treatment (4
×
6 format). The total
duration of the process was 8 h, starting at 23:00 h with the addition of refrigerated NaOH
and ending at 7:30 h with the final wash with water. The results shown for each sample
were obtained with a 1 h interval, including the final washing of the olives with water
(sample number 8 in Figure 7).
Foods 2023, 12, 2815 7 of 13
added), it is 100% (for example, after 24 h). This allows us to dynamically correct the in-
terpolator’s prediction and get a good approximation of the completion time even from
the sixth sampling.
2.6. Cloud Computing System for Commercial-Level Analysis during Production
To manage the lye treatment process remotely and at a commercial level, a web ap-
plication (lye treatment) was developed using Matlab R2022b, Glassfish 5.0, Java JDK
1.7.0, and MySQL 5.7, capable of analyzing the process online, as shown in Figure 6a,b.
(a)
(b)
Figure 6. Web application for remote management of lye treatment. (a) Flowchart of the web ap-
plication operation. (b) Login screen of the web application.
3. Results
3.1. Laboratory Tests
Figure 7 shows one of the cases studied for the lye treatment (4 × 6 format). The total
duration of the process was 8 h, starting at 23:00 h with the addition of refrigerated NaOH
and ending at 7:30 h with the final wash with water. The results shown for each sample
were obtained with a 1 h interval, including the final washing of the olives with water
(sample number 8 in Figure 7).
As can be seen, the procedure followed for obtaining the images ensures a constant
background and lighting that allows a clear distinction between the olives and the back-
ground. The contrast provided by the phenolphthalein after the drying process allows for
a perfect contrast between the area affected by NaOH and the area that has not yet been
affected.
Figure 7. The 4 × 6 format images of olives at different moments of the lye treatment process with
NaOH; the last image corresponds to the phase of washing the olives in water.
Figure 7.
The 4
×
6 format images of olives at different moments of the lye treatment process with
NaOH; the last image corresponds to the phase of washing the olives in water.
As can be seen, the procedure followed for obtaining the images ensures a constant
background and lighting that allows a clear distinction between the olives and the back-
ground. The contrast provided by the phenolphthalein after the drying process allows for
a perfect contrast between the area affected by NaOH and the area that has not yet been
affected.
In Figure 8, the images processed by the artificial vision software are shown, depicting
the temporal evolution of the percentage of lye treatment.
Foods 2023,12, 2815 8 of 13
Foods 2023, 12, 2815 8 of 13
In Figure 8, the images processed by the artificial vision software are shown, depict-
ing the temporal evolution of the percentage of lye treatment.
L.T: 13.2% L.T: 20.1% L.T: 36.2% L.T: 55.4%
L.T: 71% L.T: 77.7% L.T: 82.2% L.T: 84.3%
Figure 8. The 4 × 6 format images of olives processed with indication of the percentage of lye treat-
ment (L.T).
In each trial, a total of five repetitions are performed to obtain an average value of
the percentage of lye treatment at that determined time. The moment of finalization is
established on the basis of the type of olive and its ripeness as main factors (for this exam-
ple of an olive at the beginning of the campaign, the estimation was 85%, but it ended
with the washing water at 84.3% according to the expert’s criterion). The interpolator al-
lowed predictions of the finalization time from the first six trials with an error on the op-
timal time of finalization of the lye treatment process below 10%.
The results of the progressive use of the interpolators are shown in Figure 9, depend-
ing on the interpolation method used.
Figure 9. Interpolators used (Linear on the left, Pchip in the middle, and Makima on the right). The
figure shows the evolution in the estimation of the cooking percentage according to the sample
number.
For the case under study (see Table 1), it can be seen that the error in rounds 6 and 7
was less than 10% (bold and white) using interpolation methods such as Pchip and
Makima, both of which can serve as estimators of the moment of completion, with the
linear estimator being discarded. These data can be useful for the expert to pay attention
only to the tanks that are about to finish.
Figure 8.
The 4
×
6 format images of olives processed with indication of the percentage of lye
treatment (L.T).
In each trial, a total of five repetitions are performed to obtain an average value of
the percentage of lye treatment at that determined time. The moment of finalization is
established on the basis of the type of olive and its ripeness as main factors (for this example
of an olive at the beginning of the campaign, the estimation was 85%, but it ended with
the washing water at 84.3% according to the expert’s criterion). The interpolator allowed
predictions of the finalization time from the first six trials with an error on the optimal time
of finalization of the lye treatment process below 10%.
The results of the progressive use of the interpolators are shown in Figure 9, depending
on the interpolation method used.
Foods 2023, 12, 2815 8 of 13
In Figure 8, the images processed by the artificial vision software are shown, depict-
ing the temporal evolution of the percentage of lye treatment.
L.T: 13.2% L.T: 20.1% L.T: 36.2% L.T: 55.4%
L.T: 71% L.T: 77.7% L.T: 82.2% L.T: 84.3%
Figure 8. The 4 × 6 format images of olives processed with indication of the percentage of lye treat-
ment (L.T).
In each trial, a total of five repetitions are performed to obtain an average value of
the percentage of lye treatment at that determined time. The moment of finalization is
established on the basis of the type of olive and its ripeness as main factors (for this exam-
ple of an olive at the beginning of the campaign, the estimation was 85%, but it ended
with the washing water at 84.3% according to the expert’s criterion). The interpolator al-
lowed predictions of the finalization time from the first six trials with an error on the op-
timal time of finalization of the lye treatment process below 10%.
The results of the progressive use of the interpolators are shown in Figure 9, depend-
ing on the interpolation method used.
Figure 9. Interpolators used (Linear on the left, Pchip in the middle, and Makima on the right). The
figure shows the evolution in the estimation of the cooking percentage according to the sample
number.
For the case under study (see Table 1), it can be seen that the error in rounds 6 and 7
was less than 10% (bold and white) using interpolation methods such as Pchip and
Makima, both of which can serve as estimators of the moment of completion, with the
linear estimator being discarded. These data can be useful for the expert to pay attention
only to the tanks that are about to finish.
Figure 9.
Interpolators used (Linear on the
left
, Pchip in the
middle
, and Makima on the
right
).
The figure shows the evolution in the estimation of the cooking percentage according to the sample
number.
For the case under study (see Table 1), it can be seen that the error in rounds 6 and
7 was less than 10% (bold and white) using interpolation methods such as Pchip and
Makima, both of which can serve as estimators of the moment of completion, with the
linear estimator being discarded. These data can be useful for the expert to pay attention
only to the tanks that are about to finish.
Foods 2023,12, 2815 9 of 13
Table 1. Relative errors (%) as a function of the number of samples and interpolation methods.
n_Samples t_Lineal (h) t_Pchip (h) t_Makima
(h) t_Real (h) e_Lineal (%) e_Pchip (%) e_Makima
(%)
4 17.28 11.47 7.1 8 116 43.375 −11.25
5 14.17 9.6 6.85 8 77.125 20 −14.375
6 11.3 8.65 8.1 8 41.25 8.125 1.25
7 9.1 8.15 8.2 8 13.75 1.875 2.5
8 8.71 8.55 8.65 8 8.875 6.875 8.125
As can be seen, the final error due to the discrepancy between the predicted theoretical
percentage of completion and the one finally decided by the expert (bold and gray) was
also within 10%.
3.2. Factory Trials
The tests were carried out on 23 and 24 October 2017 in Monturque (Córdoba, Spain)
and on 2 November 2017 in Ferreira do Antalejo (Beja, Portugal). The data from 2017 were
from research trials. The data from subsequent years were internal to the olive factories and
are not publishable because they are private. The intervals between samples were those
used by the expert, and the results are shown in Table 2.
Table 2.
Relative errors (%) obtained in the Monturque tests on 23 October 2017 and 25 October 2017,
and in the Ferreira do Antalejo tests on 2 November 2017.
10/23/2017 n_Samples t_Lineal
(h) t_Pchip (h) t_Makima
(h) t_Real (h) e_Lineal
(%)
e_Pchip
(%)
e_Makima
(%)
Coc_3 7 7.42 6.71 6.67 6.92 7.23 −3.03 −3.61
Coc_4 7 7 5.35 5.04 4.75 47.37 12.63 6.11
Coc_6 7 6.1 5.73 5.95 6.59 −7.44 −13.05 −9.71
Coc_8 7 10.53 8.55 7.83 7.5 40.40 14.00 4.40
10/25/2017
Coc_5 8 7.68 6.25 6.06 5.92 29.73 5.57 2.36
Coc_8 8 10.29 9.03 9.26 8.33 23.53 8.40 11.16
Coc_9 8 7.53 6.33 6.41 6.17 22.04 2.59 3.89
Coc_10 8 6.78 5.98 5.97 6 13.00 −0.33 −0.50
11/02/2017
Coc_8 8 5.91 5.55 5.53 5.67 4.23 −2.12 −2.47
Coc_9 8 7.69 6.26 6.35 5.5 39.82 13.82 15.45
Coc_10 8 6.91 6.43 6.37 6.16 12.18 4.38 3.41
Coc_13 8 6.25 5.63 5.78 5.83 7.20 −3.43 −0.86
Average of relative error values 21.18% 6.95% 5.33%
As observed, only the error data regarding the final moment of the lye treatment
process, fixed at the expert’s discretion, are presented. Mostly, these errors were below
10% (bold) for the interpolation methods of Pchip and Makima, always overestimating the
lye treatment time for the linear case, in a ratio of 5/7 (underestimation/overestimation)
for Pchip and Makima. Underestimation can be advantageous as it alerts the expert in
advance.
A criterion that can be established is to carry out a new sample at around 50% of the
estimated finalization time after the sixth sample.
Lastly, as shown in Table 2, the average absolute error values were below 10% (bold)
for Pchip and Makima, which is a good estimate in a process that ultimately depends on
the expert’s judgment for its finalization.
Foods 2023,12, 2815 10 of 13
3.3. Use of the Cloud Computing Web Application
Figures 10–12 show the cloud computing web application. This application allows the
management of samples (including parameters such as temperature or NaOH concentration
used at the beginning of the lye treatment process), analysis of the sample evolution,
estimation of the lye treatment finishing time, management of expert personnel, and the
variety of olive used.
Foods 2023, 12, 2815 10 of 13
3.3. Use of the Cloud Computing Web Application
Figures 10–12 show the cloud computing web application. This application allows
the management of samples (including parameters such as temperature or NaOH concen-
tration used at the beginning of the lye treatment process), analysis of the sample evolu-
tion, estimation of the lye treatment finishing time, management of expert personnel, and
the variety of olive used.
New Register
Active (4)
All (4)
Finished (4)
Tanks
Tank 01
Operators
Varieties
Sign out
Update
RECORD IDENTIFIER: 7
Start of cooking: 19/06/ 2017 09:00
End of cooking: -
Tank: Tank 01
Olive variety: Manzanilla
Temperature: not available
RECORD IDENTIFIER: 6
Start of cooking: 03/06/20 17 16:23
End of cooking: -
Tank: Tank 01
Olive variety: Manzanilla
Temperature: not available
RECORD IDENTIFIER: 5
Start of cooking: 27/04/20 17 20:47
End of cooking: -
Tank: Tank 01
Olive variety: Manzanilla
Temperature: not available
RECORD IDENTIFIER: 4
Start of cooking: 09/12/20 16 18:15
End of cooking: -
Tank: Tank 01
Olive variety: Manzanilla
Temperature: not available
STATUS INFORMATION
Details of the status with identifier: 11
Delete status
Status opened on 06/19/2017 at 16:35
Percentage of lye treatmen t: 84.7
Comments
Details of the status with ident ifier: 10
Delete status
Status opened on 06/19/2017 at 14:35
Percentage of lye treatmen t: 63.1
Comments
i
Details of the status with ident ifier: 9
Delete status
Status opened on 06/19/2017 at 10:30
Percentage of lye treatmen t: 29.6
Comments
Date
Figure 10. Cloud computing application (sample management).
Status information
This lye treatment started on 04/27/2017 at 16:48
The record still remains op en
RECORD IDENTIFIER: 7
Start of cooking: 19/06/20 17 09:00
End of cooking: -
Tank: Tank 01
Olive variety: Manzanilla
Temperature: not available
RECORD IDENTIFIER: 6
Start of cooking: 03/06/ 2017 16:23
End of cooking: -
Tank: Tank 01
Olive variety: Manzanilla
Temperature: not available
RECORD IDENTIFIER: 5
Start of cooking: 27/04/20 17 20:47
End of cooking: -
Tank: Tank 01
Olive variety: Manzanilla
Temperature: not available
RECORD IDENTIFIER: 4
Start of cooking: 09/12/ 2016 18:15
End of cooking: -
Tank: Tank 01
Olive variety: Manzanilla
Temperature: not available
New status
New Register
Active (4)
All (4)
Finished (4)
Tanks
Tank 01
Operators
Varieties
Sign out
Update
i
Finish lye
treatment Edit register Delete status
Lye treatment informatio n
Tank: Tank 01
Olive variety: Manzanilla
Volume range:-
NaOH concentration:-
Temperature:-
Olive information
Texture:-
Olive Orchard:-
Pulp/Stone Ratio:-
Maturity Index:-
The record was opened by the following operator: Peter
Operator Identifier: Technician 01
Name: Peter
ID: 123456789A
Details of the status with ident ifier: 5
Status opened on 04/27/2017 at 20:47
(hours)
Figure 11. Cloud computing application (lye treatment evolution).
Figure 10. Cloud computing application (sample management).
Foods 2023, 12, 2815 10 of 13
3.3. Use of the Cloud Computing Web Application
Figures 10–12 show the cloud computing web application. This application allows
the management of samples (including parameters such as temperature or NaOH concen-
tration used at the beginning of the lye treatment process), analysis of the sample evolu-
tion, estimation of the lye treatment finishing time, management of expert personnel, and
the variety of olive used.
New Register
Active (4)
All (4)
Finished (4)
Tanks
Tank 01
Operators
Varieties
Sign out
Update
RECORD IDENTIFIER: 7
Start of cooking: 19/06 /2017 09:00
End of cooking: -
Tank: Tank 01
Olive variety: Manzanilla
Temperature: not available
RECORD IDENTIFIER: 6
Start of cooking: 03/06/ 2017 16:23
End of cooking: -
Tank: Tank 01
Olive variety: Manzanilla
Temperature: not available
RECORD IDENTIFIER: 5
Start of cooking: 27/04 /2017 20:47
End of cooking: -
Tank: Tank 01
Olive variety: Manzanilla
Temperature: not available
RECORD IDENTIFIER: 4
Start of cooking: 09/12/20 16 18:15
End of cooking: -
Tank: Tank 01
Olive variety: Manzanilla
Temperature: not available
STATUS INFORMAT ION
Details of the status wi th identifier: 11
Delete status
Status opened on 06/19/2017 at 16:35
Percentage of lye treat ment: 84.7
Comments
Details of the status with ident ifier: 10
Delete status
Status opened on 06/19/2017 at 14:35
Percentage of lye treat ment: 63.1
Comments
i
Details of the status with ident ifier: 9
Delete status
Status opened on 06/19/2017 at 10:30
Percentage of lye treat ment: 29.6
Comments
Date
Figure 10. Cloud computing application (sample management).
Status information
This lye treatment star ted on 04/27/2017 at 16:48
The record still remains op en
RECORD IDENTIFIER: 7
Start of cooking: 19/06 /2017 09:00
End of cooking: -
Tank: Tank 01
Olive variety: Manzanilla
Temperature: not available
RECORD IDENTIFIER: 6
Start of cooking: 03/06 /2017 16:23
End of cooking: -
Tank: Tank 01
Olive variety: Manzanilla
Temperature: not available
RECORD IDENTIFIER: 5
Start of cooking: 27/04/ 2017 20:47
End of cooking: -
Tank: Tank 01
Olive variety: Manzanilla
Temperature: not available
RECORD IDENTIFIER: 4
Start of cooking: 09/12 /2016 18:15
End of cooking: -
Tank: Tank 01
Olive variety: Manzanilla
Temperature: not available
New status
New Register
Active (4)
All (4)
Finished (4)
Tanks
Tank 01
Operators
Varieties
Sign out
Update
i
Finish lye
treatment Edit register Delete status
Lye treatment informatio n
Tank: Tank 01
Olive variety: Manzanilla
Volume range:-
NaOH concentration:-
Temperature:-
Olive information
Texture:-
Olive Orchard:-
Pulp/Stone Ratio:-
Maturity Index:-
The record was opened by the following operator: Peter
Operator Identifier: Technicia n 01
Name: Peter
ID: 123456789A
Details of the status with identifier: 5
Status opened on 04/27/2017 at 20:47
(hours)
Figure 11. Cloud computing application (lye treatment evolution).
Figure 11. Cloud computing application (lye treatment evolution).
Foods 2023,12, 2815 11 of 13
Foods 2023, 12, 2815 11 of 13
Staff Management
Register operator
Fields
with
operator
data
Create a variety of olive
List with
variety data
Add
ID
Operator
Identifier
Operator Identifier
Operators
Name Surname
Name: *
Surname:
ID:
Operator Identifier: *
Options
Technician 01 123456789A Peter Smith
1
1
Add
Name: *
Description:
xx
Varieties
List of olive varieties
Olive variety management
OptionsDescription
Manzanilla Test 01
Olive variety
Figure 12. Cloud computing application (staff management and olive variety management).
4. Conclusions
It is possible to quantitatively determine the evolution of olive lye treatment in NaOH
using a previous conditioning system for the olives: (1) axial cutting flush with the pit; (2)
addition of a phenolphthalein spray to obtain an intense coloration in the area affected by
NaOH; (3) drying for 5 min; (4) scanning of the olives forming a matrix in different formats
(4 × 8 or 4 × 6), where the background is replaced by a blue surface; (5) processing of the
obtained image with artificial visions.
Using a simple interpolator, it is possible to predict the optimal moment of comple-
tion of the lye treatment, which implicitly includes information about the variety, ripe-
ness, evolution of temperature during lye treatment, etc.
The interpolation method is more suitable than other options, such as a predictive
neural network, as it requires fewer data points for adjustment, unlike the latter.
The use of a web application allows an expert to simultaneously attend to several
factories and only those tanks that are near the end of the olive lye treatment process.
Although the application provides an estimate of the moment of completion of the
lye treatment, the process is constrained by the experience of the expert, who knows the
approximate times for each scenario (variety of the olive, ripeness, temperature, concen-
tration, and number of uses of the soda).
If the completion time of the lye treatment is underestimated, this errs on the side of
safety. Good practice would be to perform tastings at 50% of the time provided by the
prediction.
With a large number of lye treatments performed, it would be possible to use a neural
network for more accurate predictions or to adjust sigmoid curves.
In future work, a procedure will be tested that dispenses with the use of phenol-
phthalein and instead uses a thermal camera to profile the area penetrated by NaOH.
Author Contributions: Conceptualization, M.C.L.G., A.M.-L., J.M.M.L. and E.R.-J.; formal analysis,
M.C.L.G., A.M.-L., J.M.M.L. and E.R.-J.; investigation, M.C.L.G., A.M.-L., J.M.M.L. and E.R.-J.; writ-
ing—original draft, M.C.L.G., A.M.-L., J.M.M.L. and E.R.-J.; supervision, M.C.L.G., A.M.-L.,
J.M.M.L. and E.R.-J. All authors have read and agreed to the published version of the manuscript.
Funding: This research received no external funding.
Data Availability Statement: The data used to support the findings of this study can be made avail-
able by the corresponding author upon request.
Conflicts of Interest: The authors declare no conflict of interest.
Figure 12. Cloud computing application (staff management and olive variety management).
4. Conclusions
It is possible to quantitatively determine the evolution of olive lye treatment in NaOH
using a previous conditioning system for the olives: (1) axial cutting flush with the pit;
(2) addition of a phenolphthalein spray to obtain an intense coloration in the area affected
by NaOH; (3) drying for 5 min; (4) scanning of the olives forming a matrix in different
formats (4
×
8 or 4
×
6), where the background is replaced by a blue surface; (5) processing
of the obtained image with artificial visions.
Using a simple interpolator, it is possible to predict the optimal moment of completion
of the lye treatment, which implicitly includes information about the variety, ripeness,
evolution of temperature during lye treatment, etc.
The interpolation method is more suitable than other options, such as a predictive
neural network, as it requires fewer data points for adjustment, unlike the latter.
The use of a web application allows an expert to simultaneously attend to several
factories and only those tanks that are near the end of the olive lye treatment process.
Although the application provides an estimate of the moment of completion of the lye
treatment, the process is constrained by the experience of the expert, who knows the ap-
proximate times for each scenario (variety of the olive, ripeness, temperature, concentration,
and number of uses of the soda).
If the completion time of the lye treatment is underestimated, this errs on the side
of safety. Good practice would be to perform tastings at 50% of the time provided by the
prediction.
With a large number of lye treatments performed, it would be possible to use a neural
network for more accurate predictions or to adjust sigmoid curves.
In future work, a procedure will be tested that dispenses with the use of phenolph-
thalein and instead uses a thermal camera to profile the area penetrated by NaOH.
Author Contributions:
Conceptualization, M.C.L.G., A.M.-L., J.M.M.L. and E.R.-J.; formal analy-
sis, M.C.L.G., A.M.-L., J.M.M.L. and E.R.-J.; investigation, M.C.L.G., A.M.-L., J.M.M.L. and E.R.-J.;
writing—original draft, M.C.L.G., A.M.-L., J.M.M.L. and E.R.-J.; supervision, M.C.L.G., A.M.-L.,
J.M.M.L. and E.R.-J. All authors have read and agreed to the published version of the manuscript.
Funding: This research received no external funding.
Data Availability Statement:
The data used to support the findings of this study can be made
available by the corresponding author upon request.
Conflicts of Interest: The authors declare no conflict of interest.
Foods 2023,12, 2815 12 of 13
References
1.
Estrada Cabezas, J.M. La Aceituna de Mesa: Nociones Sobre sus Características, Elaboración y Cualidades; Diputación de Sevilla: Sevilla,
Spain, 2011.
2.
Evangelou, E.; Kiritsakis, K.; Sakellaropoulos, N.; Kiritsakis, A. Table Olives Production, Postharvest Processing, and Nutritional
Qualities. In Handbook of Vegetables and Vegetable Processing, 2nd ed.; John Wiley & Sons: Hoboken, NJ, USA, 2018; Volume 2,
pp. 727–744. [CrossRef]
3. Consejo Oleícola Internacional. Available online: https://www.internationaloliveoil.org/?lang=es (accessed on 8 May 2023).
4. Tarrado-Castellarnau, M.; Domínguez Ortega, J.M.; Tarrado-Castellarnau, A.; Pleite Gutiérrez, R. Estudio de la transferencia de
calor en la etapa de cocido en la elaboración de aceitunas verdes al estilo sevillano. Grasas y Aceites
2013
,64, 415–424. [CrossRef]
5.
Johnson, R.L.; Mitchell, A.E. Reducing Phenolics Related to Bitterness in Table Olives. J. Food Qual.
2018
,2018, 3193185. [CrossRef]
6.
Romero, C.; Brenes, M.; Yousfi, K.; García, P.; García, A.; Garrido, A. Effect of Cultivar and Processing Method on the Contents of
Polyphenols in Table Olives. J. Agric. Food Chem. 2004,52, 479–484. [CrossRef] [PubMed]
7.
Fernandez Diez, M.J.; de Castro, R.; Garrido, A.; Gonzalez Cancho, F.; Gonzalez Pelliso, F.; Nosti, M.; Heredia, A. Biotecnología de
la Aceituna de Mesa; Publicaciones CSIC: Madrid, Spain, 1985.
8.
Cano-Lamadrid, M.; Girón, I.F.; Pleite, R.; Burló, F. Quality attributes of table olives as affected by regulated de fi cit irrigation.
LWT-Food Sci. Technol. 2015,62, 19–26. [CrossRef]
9.
Rejano, L. Nuevas tendencias en el tratamiento alcalino “cocido” de las aceitunas verdes aderezadas al estilo español o sevillano.
Grasas y Aceites 2008,59, 197–204.
10. Gómez, A.H.S.; García, P.; Navarro, L.R. Elaboration of table olives. Grasas y Aceites 2006,57, 86–94.
11.
Conte, P.; Fadda, C.; Del Caro, A.; Urgeghe, P.P.; Piga, A. Table olives: An overview on effects of processing on nutritional and
sensory quality. Foods 2020,9, 514. [CrossRef]
12.
Jiménez, A.; Rodríguez, R.; Fernández-Caro, I.; Guillén, R.; Fernández-Bolaños, J.; Heredia, A. Dietary fibre content of table olives
processed under different european styles: Study of physico-chemical characteristics. J. Sci. Food Agric.
2000
,80, 1903–1908.
[CrossRef]
13.
López, A.; Montaño, A.; Garrido, A. Provitamin A carotenoids in table olives according to processing styles, cultivars, and
commercial presentations. Eur. Food Res. Technol. 2005,221, 406–411. [CrossRef]
14.
López-López, A.; Cortés-Delgado, A.; Garrido-Fernández, A. Effect of green Spanish-style processing (Manzanilla and Hojiblanca)
on the quality parameters and fatty acid and triacylglycerol compositions of olive fat. Food Chem. 2015,188, 37–45. [CrossRef]
15.
Montaño, A.; Sánchez, A.H. Chemical composition of fermented green olives. In Olives and Olive Oil in Health and Disease
Prevention; Academic Press: Cambridge, MA, USA, 2020; pp. 99–109. [CrossRef]
16.
Guo, Z.; Jia, X.; Zheng, Z.; Lu, X.; Zheng, Y.; Zheng, B.; Xiao, J. Chemical composition and nutritional function of olive (Olea
europaea L.): A review. Phytochem. Rev. 2018,17, 1091–1110. [CrossRef]
17.
Uyla¸ser, V.; Yildiz, G. The Historical Development and Nutritional Importance of Olive and Olive Oil Constituted an Important
Part of the Mediterranean Diet. Crit. Rev. Food Sci. Nutr. 2014,54, 1092–1101. [CrossRef] [PubMed]
18.
Brenes, M.; De Castro, A. Transformation of oleuropein and its hydrolysis products during Spanish-style green olive processing.
J. Sci. Food Agric. 1998,77, 353–358. [CrossRef]
19.
Diaz, R.; Faus, G.; Blasco, M.; Blasco, J.; Moltó, E. The application of a fast algorithm for the classification of olives by machine
vision. Food Res. Int. 2000,33, 305–309. [CrossRef]
20.
Diaz, R.; Gil, L.; Serrano, C.; Blasco, M.; Moltó, E.; Blasco, J. Comparison of three algorithms in the classification of table olives by
means of computer vision. J. Food Eng. 2004,61, 101–107. [CrossRef]
21.
Diaz, R. Classification and Quality Evaluation of Table Olives. In Computer Vision Technology for Food Quality Evaluation: Second
Edition; Academic Press: Cambridge, MA, USA, 2016; pp. 351–367. ISBN 9780128022320.
22.
Riquelme, M.T.; Barreiro, P.; Ruiz-Altisent, M.; Valero, C. Olive classification according to external damage using image analysis.
J. Food Eng. 2008,87, 371–379. [CrossRef]
23.
Guzmán, E.; Baeten, V.; Pierna, J.A.F.; García-Mesa, J.A. Infrared machine vision system for the automatic detection of olive fruit
quality. Talanta 2013,116, 894–898. [CrossRef] [PubMed]
24.
Ortenzi, L.; Figorilli, S.; Costa, C.; Pallottino, F.; Violino, S.; Pagano, M.; Imperi, G.; Manganiello, R.; Lanza, B.; Antonucci, F.
A machine vision rapid method to determine the ripeness degree of olive lots. Sensors 2021,21, 2940. [CrossRef]
25.
de Jódar Lázaro, M.; Luna, A.M.; Lucas Pascual, A.; Martínez, J.M.M.; Canales, A.R.; Madueño Luna, J.M.; Segovia, M.J.; Sánchez,
M.B. Deep learning in olive pitting machines by computer vision. Comput. Electron. Agric. 2020,171, 105304. [CrossRef]
26.
Pascual, A.L.; Luna, A.M.; de Jódar Lázaro, M.; Martínez, J.M.M.; Canales, A.R.; Luna, J.M.M.; Segovia, M.J. Analysis of the
functionality of the feed chain in olive pitting, slicing and stuffing machines by IoT, computer vision and neural network diagnosis.
Sensors 2020,20, 1541. [CrossRef]
27.
Macías-Macías, M.; Sánchez-Santamaria, H.; García Orellana, C.J.; González-Velasco, H.M.; Gallardo-Caballero, R.; García-Manso,
A. Mask R-CNN for quality control of table olives. Multimed. Tools Appl. 2023,82, 21657–21671. [CrossRef]
28.
Figorilli, S.; Violino, S.; Moscovini, L.; Ortenzi, L.; Salvucci, G.; Vasta, S.; Tocci, F.; Costa, C.; Toscano, P.; Pallottino, F. Olive Fruit
Selection through AI Algorithms and RGB Imaging. Foods 2022,11, 3391. [CrossRef] [PubMed]
Foods 2023,12, 2815 13 of 13
29.
Madueño Luna, A.; Pleite, R.; Madueño Luna, J.M.; Lopez Lineros, M. Procedimiento para la Determinación Cuantitativa del
Porcentaje de Cocido en Sosa Cáustica de Aceitunas y Predicción del Momento Óptimo de Finalización del Mismo. 2012, 14.
Available online: https://consultas2.oepm.es/InvenesWeb/detalle?referencia=P201100462 (accessed on 8 May 2023).
30.
Geográfica, I.; Igp, P. Pliego de Condiciones Indicación Geográfica Protegida (igp). pp. 1–15. Available online: https://www.
juntadeandalucia.es/export/drupaljda/PLIEGO_IGP_ACEITUNA_GORDAL_SEVILLANA.pdf (accessed on 21 July 2023).
31. Rejano Navarro, L. La manzanilla fina sevillana. Grasas y Aceites 1999,50, 60–66. [CrossRef]
32. Barranco, D.; Rallo, L. El Cultivo del Olivo; MundiPrensa: Madrid, Spain, 2017; ISBN 9788484767145.
33.
Consejo Oleícola Internacional. Catálogo Mundial de Variedades de Olivo; Consejo Oleícola Internacional: Madrid, Spain, 2001;
ISBN 9788493166335.
34.
Muñoz, A.C.G.; Murillo, M.S.; Albert, P.C. Catalogación y Caracterización de los Productos Típicos Agroalimentarios de Andalucía (Tomo
II); CIS Management: Amsterdam, The Netherlands, 2006; ISBN 9788495191861.
35.
Franco Rodriguez, R.; Márquez Rodriguez, V.; Soriano Castilla, I. Conocimiento Tradicional en el Olivar Sevillano; G.D.R. Serranía
Suroeste Sevillana: Marchena, Spain, 2015; ISBN 978-84-608-3953-8.
36.
Carmonaa, S.J.; De Castrob, A.; Navarrob, L.R. Proceso tradicional de aderezo de aceitunas verdes de mesa. Racionalización del
cocido. Grasas y Aceites 2011,62, 375–382. [CrossRef]
37. Lamar, D.G. Latest developments in LED drivers. Electronics 2020,9, 619. [CrossRef]
38.
Hsieh, H.I.; Wang, H. LED current balance using a variable voltage regulator with low dropout vDS control. Appl. Sci.
2017
,
7, 206. [CrossRef]
39.
Jiang, W.Z.; Hwu, K.I.; Shieh, J.J. Four-Channel Buck-Type LED Driver with Automatic Current Sharing and Soft Switching. Appl.
Sci. 2022,12, 5842. [CrossRef]
40.
Lepot, M.; Aubin, J.B.; Clemens, F.H.L.R. Interpolation in time series: An introductive overview of existing methods, their
performance criteria and uncertainty assessment. Water 2017,9, 796. [CrossRef]
41. Mathworks. Available online: https://es.mathworks.com/help/matlab/ref/interp1.html (accessed on 8 May 2023).
Disclaimer/Publisher’s Note:
The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.
