2020 International Conference on Emerging Trends in Information Technology and Engineering (ic-ETITE)

Crop Selection and IoT Based Monitoring System for

Precision Agriculture
2020 International Conference on Emerging Trends in Information Technology and Engineering (ic-ETITE) 978-1-7281-4142-8/20/$31.00 ©2020 IEEE 10.1109/ic-ETITE47903.2020.123

Yash Bhojwani Rishab Singh

SCOPE SCOPE
VIT VIT
Vellore, India Vellore, India
[email protected] [email protected]

Rachana Reddy Boominthan Perumal

SCOPE SCOPE
VIT VIT
Vellore, India Vellore, India
[email protected] [email protected]

Abstract - Internet of Things (IoT) is the future. IoT is the Smart farming is needed using IoT technologies which help
change that is required in every field. Being able to monitor and farmers to enhance crop productivity as it is a more accurate
control things from a distance makes any task effortless. and controlled process for crops. It has several applications
Agriculture is a very important field and hence every possible
technological advancement should be made in this field. With in the agricultural sector which help the farmers for a better
the rise in population worldwide, the demand for agriculture yield. It involves three major steps: smart analysis and
has increased drastically and unfortunately, farmers are failing planning, smart control and smart sensing & monitoring.
to fulfill the never-ending demand. Instead of increasing the Precision agriculture avoids unwanted wastage of raw
scale of agriculture, a better way will be implementing smart or materials. Research has proven that Precision agriculture
precision agriculture techniques using IoT. The yield of any increases the yield by 40-60% which in turn increases the
crop can be maximized with the help of precision agriculture by
reducing the wastage. The proposed work does so by monitoring premium price by 30% helping the farmers for a better
the environmental factors like temperature, humidity, soil lifestyle.
moisture, etc. that affect the growth of the crop as well as helps The current scale of agriculture is sufficient to sustain the
the farmers to decide the idle crop that will be suitable for them demand, but the amount of wastage is much higher than it
according to the data collected and environmental conditions. should be. For example, if a farm has a capacity of growing
This model can be very effective than the traditional methods as 100kgs of grains per month, typically it will produce only
the risk of crop failure, less yield, excessive water supply or
excessive use of fertilizers and pesticides, etc. can be reduced to 50kgs or 60kgs of it [12]. The reason behind less yield is
a great extent. The data collected by the sensor nodes deployed improper growth of the crop or the lack of knowledge. Many
all over the field in sent to the cloud and there the data is farmers sprinkle pesticides, water their fields and harvest
analyzed and visualized for the ease of farmers. With the help their crops according to their knowledge and idea which
of visualized data farmers can take precise and effective results in less amount of yield. Instead, if farmers adopt
decisions affecting their crops.
precision agriculture techniques using IoT, they can
maximize their yield as they would be able to analyze the
Keyword: Precision agriculture, Weather monitoring, Smart
agriculture, IoT, Thingspeak, KNN algorithm factors affecting the growth of the crop and hence reduce the
chances of crop failure, excessive or less water supply, etc.
I. INTRODUCTION Automation in the field of agriculture with the help of IoT
will help increase productivity in every stage [12].
Internet of things (IoT) simply means the connection of
We have used various sensors like DHT 11, BMP 180, SEN
various physical devices over the cloud. All the physical
0114, MQ 135, etc. to measure temperature, humidity,
devices or sensors are connected through the internet and
pressure, soil moisture, air purity, etc. These sensors will be
hence can communicate with each other. This simple idea of
connected to a microprocessor and will be installed all over
connecting physical devices over the internet to form a
the field to collect data [22]. The collected data then will be
network can be game-changing in many fields. Agriculture is
sent to the cloud with the help of an IoT hub, thingspeak.
one such field that needs technological improvements to
There the data will be analyzed with the help of MATLAB
sustain the growing food requirement and especially in
code that works in the backend of thingspeak. The data will
countries like India where crop production is much less than
then be visualized using MATLAB in real-time for the user
what is required. The traditional farming methods need to be
to understand it easily and act accordingly [23]. The user can
improved to increase crop production over the same area of
see the visualized data on his or her personal device and can
the field.
also see the predicted future values using the MATLAB code

978-1-7281-4142-8/$31.00 ©2020 IEEE

1

Authorized licensed use limited to: University of Zagreb: Faculty of Electrical Engineering and Computing. Downloaded on March 18,2021 at 08:08:22 UTC from IEEE Xplore. Restrictions apply.
 2020 International Conference on Emerging Trends in Information Technology and Engineering (ic-ETITE)

running in the backend of the website. The farmer can hence agriculture application. Three layers have been marked with
take more precise decisions seeing the values in the graph their implementation.
visualized. For example, checking the soil moisture before Having a good efficiency, this architecture can be used for
irrigating the field will help the farmer to know how much implementing different precision agriculture systems.
more water the crop requires, hence reducing the risk of Greenhouse surrounding using wireless sensor network or
excessive or low water supply that could damage the amount WMSN in the paper [6] shows that automatic irrigation is
and quality of the crop produced. more efficient than the scheduled irrigation. Automatic
Our model also helps the farmers to decide the idle crop for irrigation is successful in efficiently optimizing the usage of
farming by analyzing the temperature, climatic weight and water and fertilizers. In addition to this, it also maintains the
relative dampness that are taken by the sensors already wetness of the level of the crop at an equal level as advised
deployed for monitoring the crop. It uses KNN algorithm to by an agronomist. The authors in the paper [7] takes use of
predict which type of crop will be suitable for farming. KNN CC3200 single chip to monitor temperature and humidity in
algorithm was selected for this model as it needs no prior the agricultural field. It uses a camera to click pictures and
knowledge about the distribution of data and hence the send it to farmers mobile through multimedia messaging
training phase will also be very fast, resulting in fast results. service using Wi-Fi. Its application helps in advancing the
Also, here in this case, the already established assumptions harvesting of crops and worldwide production. Paper [8,23]
may fail making KNN more suitable for this as KNN is described the use of various technologies like cloud
nonparametric. In the end, the photo of the crop is taken and computing, big data analytics, internet of things, sensors and
analyzed using image processing and AlexNet architecture to mobile computing for creating a model for smart agriculture.
know if the crop is ready to harvest or not. Hence, this paper The model has been designed to benefit the farmers by
deals with a very cost-effective and efficient way for farmers increasing agricultural production by having a reduction in
to maximize their yield. The IoT based model implemented the cost of fertilizer requirement and cost of its Argo-
in this paper will help the farmers from selecting the crop till products. This paper [8] also helps in estimating production
the very end, harvesting it. per crop and total fertilizer requirement.
The rest of the paper is organized as follows. Section 2
describes the recent technological advancements in the III. PROPOSED WORK
agricultural field and different IoT based systems used in We propose an IoT based model to monitor and analyze the
smart agriculture, section 3 discusses about the proposed parameters affecting the growth and production of a crop in
work. Implementation and result analysis are shown in an agricultural field [10]. Our model also helps the farmers
section 4. In the end conclusion is presented and future to decide which crop will be idle for them to grow based on
improvements have been discussed in section 5. the temperature, humidity and pressure readings from the
sensors. It also helps the farmer to determine that if the crop
II. RELATED WORK is ready for harvesting or not using machine learning and
Monitoring of soil health, the environment, and managing image processing. All the sensors like temperature and
irrigation is a major challenge faced in farming. Paper [1] humidity sensor, pressure sensor, soil moisture sensor, air
discusses a system to gain control over such issues with the quality sensor, etc. are connected to the microprocessor and
help of IoT. It connects both physical sensing devices and deployed all over the field [11]. Table 1 shows all the sensors
irrigation control mechanisms to the cloud. This helps to used.
analyze the architecture. Its final results tell us that data in
Table I. Sensors used in the process
real-time can be mitigated with low latency. Migration has
shown a major hindrance to agriculture.
The author in the paper [2,14] makes use of temperature, Sensor Model
monitoring, security, moisture and GPS sensing for the
betterment of agriculture. These features can be used by users Temperature DHT11
over remote devices or the internet.
In the paper [3] a design for smart IoT communication is Humidity DHT11
presented which requires joining IoT, Wireless Sensor
Network and mapping aerial. The paper calculates the Pressure BMP180
bandwidth when different commands and data are sent by the
system. The designed system is controlled remotely using a
Soil moisture SEN0114
mobile phone. The authors of the paper [4,11] give us a
proposal of a prototype for accurate farming using a wireless
sensor network with an IOT cloud. They describe an alert Air quality MQ135
system for controlling the water stress of plants. In addition,
this application is also used to estimate the quantities of water
required. Paper [5,22] provides us with a cloud-based IoT
architecture which can be used in various precision

Authorized licensed use limited to: University of Zagreb: Faculty of Electrical Engineering and Computing. Downloaded on March 18,2021 at 08:08:22 UTC from IEEE Xplore. Restrictions apply.
 2020 International Conference on Emerging Trends in Information Technology and Engineering (ic-ETITE)

A Wi-Fi module (ESP8266) is connected to the Data acquisition from sensors, analyzing the collected data
microprocessor to send all the data collected by the sensors and visualizing the analyzed data forms the cloud server
to the cloud with the help of an IoT hub, thingspeak. Fig. 1 architecture or procedure of visualizing the data through
shows the front-end architecture or design of the model. cloud as shown in Fig. 3.
Sensors connected to the microprocessor forms the front-end
node of the model. There will be many front-end nodes
deployed all over the field collecting and sending data to the
cloud via internet [9].

Fig. 3. Cloud server architecture

The model can also estimate future values based on

previously collected data by the sensors stored in the database
using predictive analysis with the help of MATLAB running
Fig. 1. Front-end node architecture in the backend. System architecture and flow of the IoT based
monitoring system for precision agriculture is depicted in
Using thingspeak ensures data security, real-time data Fig. 4.
visualization, and many added advantages. It uses MQTT
protocol which is idle for IoT applications as it is highly
scalable, lightweight, minimizes network bandwidth and
very secure.
The data collected is then analyzed and visualized in the form
of graphs with the help of MATLAB running in the backend
of thingspeak [19]. The visualized data can be directly
viewed on any device like a tablet, computer or phone from
any location. As the data is already visualized, farmers can
easily understand it and take accurate decisions based on the
data. Fig. 2 depicts working of an IoT platform and remote
users accessing the cloud via internet using any device.

Fig. 2. IoT platform Fig. 4 System architecture of the monitoring system

Authorized licensed use limited to: University of Zagreb: Faculty of Electrical Engineering and Computing. Downloaded on March 18,2021 at 08:08:22 UTC from IEEE Xplore. Restrictions apply.
 2020 International Conference on Emerging Trends in Information Technology and Engineering (ic-ETITE)

The process starts with the data acquisition from If the conditions are not idle then actions can be taken
environment, soil or agricultural field and plant. All this data accordingly and then check the values again on thingspeak.
is then sent to thingspeak with the help of a Wi-Fi module The data is sent in real time and hence any change can be
connected to the microprocessor [13]. There the data is viewed instantly. Same procedure can be continued till the
processed and visualized. Seeing the visualized data farmers end. Fig. 5 represents the flow of the proposed system used
can decide if the conditions are idle for the growth of crop. to monitor the factors affecting growth of the crop.

Fig. 5. Flow Diagram of the proposed system

Authorized licensed use limited to: University of Zagreb: Faculty of Electrical Engineering and Computing. Downloaded on March 18,2021 at 08:08:22 UTC from IEEE Xplore. Restrictions apply.
 2020 International Conference on Emerging Trends in Information Technology and Engineering (ic-ETITE)

By taking the temperature, humidity, and atmospheric Fig. 6. Block Diagram of crop selection architecture
pressure values as inputs, the model can predict the idle crop KNN algorithm is used as it is non-parametric and hence
for cultivation using KNN (K-Nearest Neighbors) algorithm. can work with a very small distributed data set which is
Fig. 6 shows the block diagram of the system and the sensors very important for our model. Also, the training phase is
required to find the idle crop using Raspberry Pi. The data is very minimal, which means the training phase will be very
stored on the thingspeak database and can be viewed on the fast. The output or which crop will be idle to cultivate is
web app connected with thingspeak. By entering the IP based on feature similarity. This means that the crop whose
address of the website on any browser, the user can see the features will have the maximum similarity with the data
results at any remote location. point based on the training set will be selected [15].

After taking all the required values and sending the data to
thingspeak, a Node JS server is used to connect
thingspeaks.com website database to our web application.
Now, the values of the sensors can be directly viewed on the
web app by just entering the URL. User can use any device
to see the results on the web app. Fig. 7 represents a flow
diagram the shows how the data is taken from the sensors and
displayed on the web app.

Fig. 6. Block diagram of selection architecture

Fig. 7. Flow Diagram for determining the idle crop

Authorized licensed use limited to: University of Zagreb: Faculty of Electrical Engineering and Computing. Downloaded on March 18,2021 at 08:08:22 UTC from IEEE Xplore. Restrictions apply.
 2020 International Conference on Emerging Trends in Information Technology and Engineering (ic-ETITE)

Our model also tells if the crop is ready for harvesting. Image The data is aggregated, analyzed and visualized with the help
of the crop is taken as input and statistical features are of MATLAB working in the backend of thingspeak. MQTT
extracted from the segmented grayscale image of the crop in protocol is used to store and retrieve data from sensors over
wavelet domain using Haar filter. Depending on the color of the internet making the data transfer secure and fast. It
the crop, masking is used to determine how much part of the enables users to see the visualized data in real time as shown
crop is mature. Many crops have different color before and in the following graphs. As the user can have real time data
after being matured or when ready to harvest. This is very visualization, actions can be taken based on the data and
important and useful while image masking. Different filters changes in the values can be noticed after the action.
and colors for masking can be applied depending on the crop.
After the image processing part, a data set is generated or The graph in Fig. 10 represents the temperature change with
taken for the crop. Then according to the data set a trainer respect to time. DHT11 is used to measure the temperature
code is used to train the model and finally a detector code to and the data is sent to thingspeak for visualization. User can
detect the level of maturity of the crop or when the crop is see the temperature variations in the graph and act
ready to harvest using image processing. Parameters and data accordingly.
set will be different for every crop [16].

IV. IMPLEMENTATION AND RESULT ANALYSIS

User can see the visualized data on thingspeak. Data will be
represented in the form of graphs and for each feature that
affects the growth of the crop, a separate graph will be
visualized with respect to time.
All the sensors are connected to the microprocessor and using
Wi-Fi module data is sent to thingspeak. Fig. 8 shows the
connection of temperature sensor to the microprocessor. The
data sensed is sent to thingspeak using ESP8266.
Similarly, Fig. 9 shows the setup for checking the air quality.

Fig. 10. Analysis of Temperature value

The graph in Fig. 11 represents the humidity percentage with

respect to time. DHT11 is used to measure the humidity
percentage

Fig. 8. Experimental setup for DHT11 sensor

Fig. 11. Analysis of Humidity value

Fig. 9. Experimental setup for checking the air quality

Authorized licensed use limited to: University of Zagreb: Faculty of Electrical Engineering and Computing. Downloaded on March 18,2021 at 08:08:22 UTC from IEEE Xplore. Restrictions apply.
 2020 International Conference on Emerging Trends in Information Technology and Engineering (ic-ETITE)

The graph in Fig. 12 represents the soil moisture percentage The graph in Fig. 14 represents the atmospheric pressure
with respect to time. SEN0114 is used to measure the soil with respect to date. BMP180 is used to measure the
moisture percentage. Depending on the soil moisture atmospheric pressure.
percentage, farmers can take decision to irrigate the field.

Fig. 12. Analysis of Soil Moisture value Fig. 14. Analysis of Pressure value

The graph in Fig. 13 represents the air quality value with The graph in Fig. 15 represents the carbon dioxide
respect to time. MQ135 is used to measure the air quality concentration with respect to date.
value. Bad air quality can affect the crops growth and
quality. Hence air quality should be maintained and
monitored.

Fig. 13. Analysis of Air Quality value Fig. 15. Analysis of Carbon dioxide value

Authorized licensed use limited to: University of Zagreb: Faculty of Electrical Engineering and Computing. Downloaded on March 18,2021 at 08:08:22 UTC from IEEE Xplore. Restrictions apply.
 2020 International Conference on Emerging Trends in Information Technology and Engineering (ic-ETITE)

Fig. 16 shows the channel location. Knowing the location

can be helpful in analyzing the type of soil in that area and
can also help in selecting the crop for cultivation.

Fig.18. Output of temperature, humidity and atmospheric pressure

Fig. 19, Fig. 20 and Fig.21 shows the values of the three
parameters required to determine the idle crop for cultivation
on the web application. The web application is made minimal
keeping in mind the end users will mostly be farmers and
many of them can have difficulties in using a complex
interface.

Fig. 16. Thingspeak Channel location

For determining the crop only DHT11 and BMP180 is

required. Fig. 17 shows the setup required for crop selection.
The data is being displayed on the web app and according to
the data idle class for crop is selected.
Fig. 19. Temperature reading on web app

Fig. 20. Pressure reading on web app

Fig. 17. Experimental setup for crop selection using DHT11 and BMP180

Fig. 18 shows the value of temperature, humidity and

atmospheric pressure in python. This data can be displayed
on the web application for better view and also can be used
for determining the idle crop for cultivation using the KNN
algorithm. These values will be taken as inputs for the
algorithm. Result will be based on these three parameters
only. Fig. 21. Humidity reading on web app

Authorized licensed use limited to: University of Zagreb: Faculty of Electrical Engineering and Computing. Downloaded on March 18,2021 at 08:08:22 UTC from IEEE Xplore. Restrictions apply.
 2020 International Conference on Emerging Trends in Information Technology and Engineering (ic-ETITE)

For crop selection temperature, humidity and pressure values The last step is to check if the crop is mature or ready to
are taken as input and based on the training data file crop is harvest. In order to test our system, we chose banana as a
selected. Fig. 22 shows the predicted class according to the crop. We chose the color for masking and the filter according
sample. Training data file used to train the model is iris- to the crop, banana. An image is taken as the input of the
dataset and testing data file to test the accuracy is named iris- sample crop banana. The output shows on a scale of 1-10 that
test. Iris-Sentosa, Iris-versicolor and Iris-virginica are the how ripe the banana is. Fig. 23 shows the image of a ripe
names of class (crop). One of them is chosen according to the banana as an input and output of the image is ‘Very Ripe –
sample. 7+’, which means the crop is ready to harvest. The result of
the program matches with the already known result.

Fig. 22. Crop selection on the basis of temperature, humidity and pressure values from web app

Fig.23. Maturity of the sample crop banana

Authorized licensed use limited to: University of Zagreb: Faculty of Electrical Engineering and Computing. Downloaded on March 18,2021 at 08:08:22 UTC from IEEE Xplore. Restrictions apply.
 2020 International Conference on Emerging Trends in Information Technology and Engineering (ic-ETITE)

V. CONCLUSION AND FUTURE WORK [5] Khattab, A., Abdelgawad, A., & Yelmarthi, K.
Agriculture is one of the most important things in any country (2016, December). Design and implementation of a cloud-
and its improvement is very vital. Traditional farming based IoT scheme for precision agriculture. In 2016 28th
methods are not very efficient and farmers usually face a lot International Conference on Microelectronics (ICM) (pp.
of problems that cannot be resolved by traditional methods. 201-204). IEEE.
Hence technological improvement in this field is necessary. [6] Mat, I., Kassim, M. R. M., Harun, A. N., & Yusoff,
Smart agriculture or precision agriculture using IoT is a very I. M. (2016, October). IoT in precision agriculture
efficient and effective method compared to the traditional applications using wireless moisture sensor network. In 2016
ones. In this paper, we demonstrated an IoT based model for IEEE Conference on Open Systems (ICOS) (pp. 24-29).
smart agriculture that is capable of solving many problems IEEE.
that farmers face. Our model has three modules- crop [7] Prathibha, S. R., Hongal, A., & Jyothi, M. P. (2017,
selection module, crop monitoring system module and crop March). IOT Based monitoring system in smart agriculture.
maturity level module. All three modules are properly In 2017 International Conference on Recent Advances in
described in the paper with the architecture of each module Electronics and Communication Technology (ICRAECT)
and our result shows that the proposed work is very useful in (pp. 81-84). IEEE.
making accurate decisions based on the visualized data [8] Channe, H., Kothari, S., & Kadam, D. (2015).
viewable at the user end through a Web app. All the data Multidisciplinary model for smart agriculture using internet-
collected by different sensors is analyzed and visualized in of-things (IoT), sensors, cloud-computing, mobile-
the form of graphs. Based on the values farmers can take computing & big-data analysis. Int. J. Computer
Technology & Applications, 6(3), 374-382.
decisions like if the soil moisture percentage is less, crops can
[9] Yoon, C., Huh, M., Kang, S. G., Park, J., & Lee, C.
be irrigated. The overall cost of the system is very less and
(2018, February). Implement smart farm with IoT
the sensors used are easily available. Hence this system can
technology. In 2018 20th International Conference on
be used by every farmer to increase their crop production.
Advanced Communication Technology (ICACT) (pp. 749-
Many more improvements can be made in the proposed
752). IEEE. [10] Gondchawar, N., & Kawitkar, R. S. (2016).
system for the future development of the model. An alarm or
IoT based smart agriculture. International Journal of
LED can be integrated with this system to notify warnings
advanced research in Computer and Communication
regarding temperature, humidity, air quality, etc. More
sensors like PIR sensor, leaf wetness sensor, etc. can be Engineering, 5(6), 838-842.
added for additional functionalities. The crop maturity level [11] Patil, K. A., & Kale, N. R. (2016, December). A
system can be integrated with a simple drone having a camera model for smart agriculture using IoT. In 2016 International
to cover a large agricultural field and to know how much of Conference on Global Trends in Signal Processing,
it is ready to be harvested. Machine learning algorithms can Information Computing and Communication
be used for predictive analysis running in the backed for (ICGTSPICC) (pp. 543-545). IEEE.
farmers to know about the climatic conditions in advance. All [12] Gayatri, M. K., Jayasakthi, J., & Mala, G. A. (2015,
these improvements will lead to an increased crop production July). Providing Smart Agricultural solutions to farmers for
and will reduce the amount of crop failure compared to the better yielding using IoT. In 2015 IEEE Technological
traditional methods. Innovation in ICT for Agriculture and Rural Development
(TIAR) (pp. 40-43). IEEE.
[13] Yuanguai, L., Long, W. A. N. G., & Jiang, W. U.
VI. REFERENCES (2009). Application of Zigbee wireless sensor network in
[1] Jaishetty, S. A., & Patil, R. (2016). IoT sensor Precision agriculture [J]. Journal of Qiongzhou University, 5,
networkbased approach for agricultural field monitoring and 32-34. [14] Verdouw, C., Wolfert, S., & Tekinerdogan, B.
control. IJRET: International Journal of Research in (2016). Internet of Things in agriculture. CAB Reviews:
Engineering and Technology, 5(06). Perspectives in Agriculture, Veterinary Science, Nutrition
[2] Suma, N., Samson, S. R., Saranya, S., and Natural Resources, 11(35), 1-12.
Shanmugapriya, G., & Subhashri, R. (2017). IOT based [15] Veenadhari, S., Misra, B., & Singh, C. D. (2014,
smart agriculture monitoring system. International Journal January). Machine learning approach for forecasting crop
on Recent and Innovation Trends in computing and yield based on climatic parameters. In 2014 International
communication, 5(2), 177-181. Conference on Computer Communication and
[3] Cambra, C., Sendra, S., Lloret, J., & Garcia, L. Informatics (pp. 1-5). IEEE.
(2017, May). An IoT service-oriented system for agriculture [16] Khera, N., Shukla, D., Sharma, P., & Dar, I. G.
monitoring. In 2017 IEEE International Conference on (2017, September). Development of microcontroller based
Communications (ICC) (pp. 1-6). IEEE. lemon ripening condition monitoring system. In 2017 6th
[4] Karim, F., & Karim, F. (2017). Monitoring system International Conference on Reliability, Infocom
using web of things in precision agriculture. Procedia Technologies and Optimization (Trends and Future
Computer Science, 110, 402-409. Directions)(ICRITO) (pp. 591-593). IEEE.

10

Authorized licensed use limited to: University of Zagreb: Faculty of Electrical Engineering and Computing. Downloaded on March 18,2021 at 08:08:22 UTC from IEEE Xplore. Restrictions apply.
 2020 International Conference on Emerging Trends in Information Technology and Engineering (ic-ETITE)

[17] Jhuria, M., Kumar, A., & Borse, R. (2013,

December). Image processing for smart farming: Detection
of disease and fruit grading. In 2013 IEEE Second
International Conference on Image Information Processing
(ICIIP-2013) (pp. 521-526). IEEE.
[18] Gutiérrez, J., Villa-Medina, J. F., Nieto-Garibay, A.,
& Porta-Gándara, M. Á. (2013). Automated irrigation system
using a wireless sensor network and GPRS module. IEEE
transactions on instrumentation and measurement, 63(1),
166-176.
[19] Mohanraj, I., Ashokumar, K., & Naren, J. (2016).
Field monitoring and automation using IOT in agriculture
domain. Procedia Computer Science, 93, 931-939. [20]
Nandurkar, S. R., Thool, V. R., & Thool, R. C. (2014,
February). Design and development of precision agriculture
system using wireless sensor network. In 2014 First
International Conference on Automation, Control, Energy
and Systems (ACES) (pp. 1-6). IEEE.
[21] Nikolidakis, S. A., Kandris, D., Vergados, D. D., &
Douligeris, C. (2015). Energy efficient automated control of
irrigation in agriculture by using wireless sensor networks.
Computers and Electronics in Agriculture, 113, 154-163.
[22] TongKe, F. (2013). Smart agriculture based on
cloud computing and IOT. Journal of Convergence
Information Technology, 8(2).
[23] Roy, S., Ray, R., Roy, A., Sinha, S., Mukherjee, G.,
Pyne, S., ... & Hazra, S. (2017, August). IoT, big data science
& analytics, cloud computing and mobile app based hybrid
system for smart agriculture. In 2017 8th Annual Industrial
Automation and Electromechanical Engineering Conference
(IEMECON) (pp. 303-304).
IEEE.

11

Authorized licensed use limited to: University of Zagreb: Faculty of Electrical Engineering and Computing. Downloaded on March 18,2021 at 08:08:22 UTC from IEEE Xplore. Restrictions apply.