=Paper=
{{Paper
|id=Vol-2761/HAICTA_2020_paper19
|storemode=property
|title=Application of RFID Technology for Evaluation the Weather Condition in Soybean Sown on Different Row Spacing
|pdfUrl=https://ceur-ws.org/Vol-2761/HAICTA_2020_paper19.pdf
|volume=Vol-2761
|authors=Denis Stajnko,Miran Lakota
|dblpUrl=https://dblp.org/rec/conf/haicta/StajnkoL20
}}
==Application of RFID Technology for Evaluation the Weather Condition in Soybean Sown on Different Row Spacing==
https://ceur-ws.org/Vol-2761/HAICTA_2020_paper19.pdf
Application of RFID Technology for Evaluation the
Weather Condition in Soybean Plant Sown on Different
Row Spacing
Denis Stajnko1, Miran Lakota2
1
University of Maribor, Faculty of Agriculture and Life Sciences, Chair for biosystem
engineering, Pivola 10, SI-2311 Hoce, Slovenia; email: denis.stajnko@um.si
2
University of Maribor, Faculty of Agriculture and Life Sciences, Chair for biosystem
engineering, Pivola 10, SI-2311 Hoce, Slovenia
Abstract. The effect of different row spacing (35 cm and 70 cm) on the
development of soybean plants was studied using two soybean varieties, Lenka
and ES Mentor. By using RFID technology, which includes a smart tag with a
microchip, temperature, moisture and three sensors, an antenna and a reader, we
measured temperature, humidity and light in the middle of the row space of each
experimental field throughout the growing season. Besides, crop development,
measurements of the morphological characteristics of the plants inventory of
weeds and yield weight were made. We discovered that differences between
soybean varieties were more pronounced than differences between different row
spacing. The application of RFID technology was proved to be accurate enough
in plant growing, so together with proposed improvements RFID would allow
the farmer to control agricultural land more quickly and to respond in a shorter
amount of time in comparison state meteorological station.
Keywords: RFID; soybeans; row spacing; weather data.
1 Introduction
RFID (Radio frequency identification) is an emerging technology that can have
enormous opportunities in Agri-food sector. Technological developments in the area
of networking devices, sensors and communication technology play significant role in
sustainability of Agri-food sector. RFID is one of such pervasive technology, which is
now increasingly utilized in food logistics, supply chain management, cold chain
monitoring and retail (Ruiz-Garcia and Lunadei, 2011). Compared with traditional
technologies like, barcode and data loggers, RFID shows several advantages. RFID
devices do not require visual contact (Abad et al., 2009). They can be placed inside
boxes, containers, embedded in any object or injected into animals (Finkenzeller,
2004). RFID tags can write 5 tags per second and have storage capacity: 1 MB in active
tags and 4 kB in passive (EPC Global, 2008). RFID tag with associated hardware and
software provides additional benefits such as real time monitoring, environmental
sensing, tracing and tracking. Different types of sensor can be embedded with RFID
114
Tags to collect information about various parameters like temperature (Amador and
Emond, 2010), humidity (Abad et al., 2009), and light intensity (Abad et al., 2009).
Biosensor tags are also under development that could be used for detecting bacterial
contamination of food products along the supply chain (Wentworth, 2003). The main
purpose of our research was to use RFID technologies with sensors for measuring
temperature, humidity in light intensity between soybean plants during the vegetation.
Two varieties of soybeans ‘Lenka’ and ‘ES Mentor’ both of the same maturity class
(00 were grown on 35 cm and 70 cm row spacing. We investigated the differences in
plant development as well as temperature, humidity and light gradients between
varieties and row spacing. Finally, we estimated the accuracy of RFID with the
measurement of meteorological conditions on nearby meteorological station.
2 Material and Methods
The soybean experiment was carried out on the field owned by a local farmer, it is
located in the municipality of Hoče-Slivnica (Slovenia). The plot ‘Štuk’ (46o30’28’’N,
15o039’19’’E) has 2.37 ha in area with the average altitude of 272 m and the average
slope is 1% and 0.6 °, respectively.
For the purpose of the experiment, the south-eastern part of the parcel was selected
and divided into four experimental sub-parcels with 1200 m2 each (Fig. 1). The fields
were visibly divided by 0.75 m wide "border bands", where no crop was presented.
The relatively large area of the experimental fields allowed for a more precise choice
of location for the placement of the measuring devices and represented a more
representative sample for the analysis of the results.
The prerequisite of soybeans was winter wheat, followed by white mustard after
harvest. The experimental boxes were separated by wooden stakes, marked with
variety and row spacing.
Fig. 1. Location of trial plots.
115
2.1 Description of Devices
We used a measuring system consisting of a semi-passive smart tag with an
integrated circuit, in our case it was a SL900A microchip, which has an integrated
temperature sensor and allows the connection of additional external sensors.
Furthermore, the measuring system consists of sensors, antenna and battery. We
collectively call this set a smart badge or Tag (Fig. 2). The measuring system also
includes an antenna reader, in our case, the DRM-900A, which transfers data from a
badge to a computer. The smart badge was equipped with a humidity sensor (EMD-
4000) and an illumination sensor (TEMD6200FX01). The described RFID measuring
system automatically measures and stores data every 2 hours. In this way, the memory
was sufficient for 14 days of data storage. Battery life is up to five years, depending
on weather conditions and measurement time interval. The data transmission range can
be up to 100 m with a properly performing antenna, and the transmission starts when
the radio frequency signal of the reader is received. The system operates at ultra-high
frequencies (UHF 860-960 MHz) according to the master-slave principle.
Fig. 2. Smart badge with microchip, sensors, antenna and battery button.
116
2.2 Optical Sensor TEMD6200FX01
The sensor is a semiconductor element - a closed-polarized diode that converts light
into electrical current. Sensor size is 2mm x 1.25mm x 0.85mm. The spectral
sensitivity of the diode is adapted to the response of the human eye. EMD-4000
Moisture Sensor The sensor substance is a special polymer deposited on a bismuth
coated ceramic substrate. The polymer changes the impedance in the presence of
moisture. The smart badge excites an alternating current sensor and measures
impedance as a function of relative humidity. The size of the sensor is 10 mm x 5 mm
x 1.5 mm.
3 Results
3. 1 Comparison of morphological properties
As seen from Fig. 4 the highest average plant height was measured by ‘Lenka’ 35
cm (110.1 cm). The highest number of branches per plant was recorded with ‘EC
Mentor’ 35 cm (57). The maximum distance of the first branch from the ground (13.5
cm) was measured by both varieties at a distance of 35 cm between the rows. The
‘Lenka’ variety had more shoots that are lateral at both interspace distances than the
‘ES Mentor’ variety.
Table 1. Comparison of physiological characteristics between ‘Lenka’ and ’ES Mentor’ July
30, 2019.
Treatment Height Number Distance from Number of
(cm) of pods soil (cm) buds
‘Lenka’ 35 cm 110.1a 43.8b 13.5a 13.7a
‘Lenka’ 70 cm 108.4a 45.9b 12.8b 13.4a
‘ES Mentor’ 70 cm 92.7b 52.8a 11.9c 10.2b
‘ES Mentor’ 35 cm 92.7b 57.2a 13.4a 11.9b
a, b
sig. at p<0.05 Duncan test
3. 2 Measurements of temperature with RFID
Table 2 shows the average values of the individual sensor at each of three
measurement intervals as well as the reference values obtained from the closest ARSO
meteorological station Edvard Rusjan Maribor Airport (ARSO, 2019) located 850 m
from our field. The data measured on the experimental plots of the ‘Lenka’ 35 cm in
all intervals and ‘ES Mentor’ 35 cm in the second and third measuring interval
correspond with the data obtained by ARSO, while all the other measurements were
significantly lower. The reason for differences is the coverage by the soybean plants
in those particular variants. Comparison between varieties for the ‘ES Mentor’ 35 cm
117
variety shows a bigger coverage of the row space at a distance of 70 cm. Despite the
average number of buds per plant of the ‘Lenka’ variety, this covered less row space.
‘Lenka’ 70 cm covered an average of 59.6% of row space, while the ’ES Mentor’ 70
cm averaged 70.7% of row space, respectively.
Table 2. Temperature measurements with RFID.
Average temperature [°C]
Treatment June 30-July 14 July 15-July 30 July 31-Aug 14
ARSO 20.93b 22.65a 22.96a
‘Lenka’ 35 cm 22.10a 22.98a 22.67a
‘Lenka’ 70 cm 18.29b 19.24b 19.33b
‘ES Mentor’ 70 cm 18.93b 20.62b 19.67b
‘ES Mentor’ 35 cm 18.04b 22.11a 22.06a
a, b
sig. at p<0.05 Duncan test
3. 3 Measurements of humidity with RFID
Table 3 shows the average values of humidity for the individual sensors at each
three measurement intervals. The data measured correlate opposed as the temperature
development due to the differences in coverage between varieties and row spacing.
Thus, the highest average relative humidity was measured in ’ES Mentor’ 35 cm but
did significantly differ only from ‘Lenka’ 35 cm as well as from the data of
meteorological station. Later shows smaller values due to the open meadow in which
the data very captured.
Table 3. Humidity measurements with RFID.
Average humidity [%]
Treatment June 30-July 14 July 15-July 30 July 31-Aug 14
ARSO 70.12a 68.95a 70.40a
‘Lenka’ 35 cm 69.10a 67.98a 70.67a
‘Lenka’ 70 cm 72.29b 70.24b 72.89b
‘ES Mentor’ 70 cm 72.93b 70.62b 72.97b
‘ES Mentor’ 35 cm 72.04b 70.11a 73.06a
a, b
sig. at p<0.05 Duncan test
3. 4 Measurements of light intensity with RFID
As seen from Table 4 the light intensity is significantly lower below the leaves in
the zone of sensors in all varieties and row spacing. Additionally, from the first to the
last measuring period the development of leaves decreased the light intensity
significantly in the same treatment. However, in the second and third period the light
intensity was significantly lower in ‘ES Mentor’ comparing to ‘Lenka’ in both row
spacing. This corresponds also with the higher number of pods in ‘ES Mentor’ (Table
1).
118
Table 4. Light intensity measurements with RFID.
Average light intensity [lx]
Treatment June 30-July 14 July 15-July 30 July 31-Aug 14
ARSO 28500a 29400a 30800a
‘Lenka’ 35 cm 9300b 6400b 4300b
‘Lenka’ 70 cm 10050b 6100b 4500b
‘ES Mentor’ 70 cm 8800b 3700c 3300c
‘ES Mentor’ 35 cm 7800c 4400c 2800c
a, b,c
sig. at p<0.05 Duncan test
3. 5 Yield of soybean
On October 13 the soybean was harvested and weighed directly on the field. It
contained 12.5 % of moisture so later it had to be additionally dried out to storage
moisture of 9 %. From Table 5, showing the yield of soybean with 9 % moisture, we
can see that the ‘ES Mentor’ had significantly higher yield than ‘Lenka’ in both row
spacing. On the other hand, there was no difference between the row spacing among
the same variety, either.
Table 5. Yield of soybean with 9 % moisture
Treatment Yield [kg/ha]
‘Lenka’ 35 cm 2367b
‘Lenka’ 70 cm 2521b
‘ES Mentor’ 70 cm 4637a
‘ES Mentor’ 35 cm 4512a
a, b
sig. at p<0.05 Duncan test
4 Conclusions
The effect of different row spacing of two soybean varieties on development of
plants was studied using RFID technology to measure temperature, light and humidity
in the middle of row spacing. Since the growing phase R1, we estimated the height of
the plants, the distance of the lower pods from the ground, the number of lateral buds
and the number of pods per plant. We found that the row spacing did not have a
significant effect on the development of the crop within the same variety. However,
the biggest differences were observed between the two soybean varieties.
119
RFID showed to be a suitable tool for accurate measurements of temperature and
humidity more accurately than the official meteorological station, which might serve
for predicting the start of many fungal diseases on time.
The UHF-RFID technologies have also some disadvantages. Above all, the problem
is the reliability of the connection, since electrically conductive materials represent a
barrier to electromagnetic waves (reflection and wave absorption). Thus, moisture in
plants is also a barrier. These problems do not have a Magnetic Assembly (HF) based
badge.
The second problem represents the storage capacity of RFID as well as the reading
and transferring of data every few weeks. The easiest way would be to read data from
RFID in the field using a drone with a built-in UHF reader. The drone would be
connected to a home computer via a WiFi network, which would determine, for
example, the time, method or means of spraying, based on the built-in algorithms and
data collected. If such a drone were equipped with hyperspectral imaging cameras,
then algorithms could be installed to separate and detect the state of plants based on
spectral signatures. This would give a reliable prediction of the necessary measures,
which would be an effective contribution to the future vision of technologically
advanced farms and smart villages.
Acknowledgement. This article was partly created as a result of the applied project
‘Reducing of draught stress and increasing of soil fertility by introducing conservation
(conservation) soil tillage into sustainable agriculture (V4-1815)’ funded by the Public
Research Agency of the Republic of Slovenia and the Ministry of Agriculture. Forestry
and Food of the Republic of Slovenia. Authors also thank to Mišel Pleteršek for his
vital contribution in managing practical experiment in his field.
References
1. Abad, E., Palacio, F., Nuin, M., Zárate, A. G. d., Juarros, A. and Gómez, J. M.
(2009). RFID smart tag for traceability and cold chain monitoring of foods:
Demonstration in an intercontinental fresh fish logistic chain. Journal of Food
Engineering, 93, 4, p.394-399.
2. AgPhD. 2019. (http://www.agphd.com, accessed on September 3, 2019)
3. Alves, B.J.R., Boddey, R.M. and Urquiaga, S. (2003). The success of BNF in
soybean in Brazil. Plant Soil, 252: p.1-9.
4. Amador, C. and Emond, J. P. (2010). Evaluation of sensor readability and thermal
relevance for RFID temperature tracking. Computers and Electronics in
Agriculture, 73, 1, p.84-90.
5. ARSO. (2019) Vreme podrobneje. 2019. http://meteo.arso.gov.si/met/sl/ (accessed
on September 11, 2019).
6. Finkenzeller, K. (2004). Radio-Frequency Identification Fundamentals and
Applications. RFID (2nd ed.).In Handbook, England: John Wiley and Sons.
120
7. Isoda, A. and Komaki, K. (2018). Varietal Differences in Flowering, Pod Setting
and Photosynthesis in Soybean Under High Temperature Conditions, Advances in
Crop Science and Technology, Adv Crop Sci Tech, 6(5), p. 392.
8. EPC Global. (2008). EPCTM radio-frequency identity protocols Class-1
Generation-2 UHF RFID protocol for communications at 860 MHzw 960MHz.
9. Lenka, https://www.agrosaat.si/sorte-hibridi-gnojila-fitofarmacevtska-
sredstva/lenka/ (accessed on September 15, 2019).
10. Mulla, D.J. (2012). Twenty-five years of remote sensing in precision agriculture:
Key advances and remaining knowledge gaps, Biosystems engineering, 114,
p.358-371.
11. Muralidhara, B. L. and Geethanjali B. (2018). A Review on different technologies
used in Agriculture. International Journal of Pure and Applied Mathematics, 119,
16, p. 4117-4134.
12. Pedersen, P. (2019). Row Spacing in Soybean. Iowa State University, Dept. Of
Agronomy. https://crops.extension.iastate.edu/files/article/RowSpacing_000.pdf
(accessed on September 15, 2019)
13. Ruiz-Garcia, L., and Lunadei, L. (2011). The role of RFID in agriculture:
Applications, limitations and challenges. Computers and Electronics in
Agriculture, 79, 1, p.42-50.
14. Wallach, D. and Thorburn, P.J. (2017). Estimating uncertainty in crop model
predictions: Current situation and future prospects, European Journal of
Agronomy, 88, chapter A1-A7, Elsevier.
121