an iot based innovative irrigation model for smart and ...tierarztliche.com › gallery ›...

An IoT based Innovative Irrigation Model for Smart and

Precision Agriculture

P.Suresh1,S.Koteeswaran2, R.H.Aswathy3,N. Malar vizhi4 , C.Balamurugan5

1 Research scholar,

Veltech Rangarajan Dr. Sagunthala R & D Institute of science and Technology, 1Assistant professor,

KPR Institute of Engineering and Technology, Coimbature, India

[[email protected]] 2 Associate Professor,

Veltech Rangarajan Dr. Sagunthala R & D Institute of science and Technology,

3Assistant professor,

KPR Institute of Engineering and Technology, Coimbature, India

[[email protected]] 4 Professor,

Veltech Rangarajan Dr. Sagunthala R & D Institute of science and Technology,

Chennai, India

[[email protected]] 5Assistant Professor,

VV College of Engineering,

Tirunelveli, India

[[email protected]]

Abstract

The current situation of agriculture is unsatisfactory for its production.The farmers

are helpless to bring the maximum yield in agriculture due to the lack of recent advances in the

technology. The contribution of technological advancement in agriculture is less attentive in

many developing nations. Applying and working with new technology in agriculture to maximize

the yield is the major issue because of the literacy of the farmers. Internet of Things (IoT), as a

breaking and newly evolving technology, attracts and helps to the productivity in agriculture.

The birth of the Internet of Things (IoT) brings revitalization in the agricultural domain. IoT

helps in forecasting crop yield and other factors promoting the good yield. The factors like the

temperature of the soil, real-time observations on air quality, level of water and suitable timing

of delivering the crop to market and so on are sensed and measures with the help of components

of the IoT framework to increase productivity. In this research work, an innovative model named

“Precision Agricultural Recommendation Model (PARM)” is proposed for boosting the yield of

agriculture. This model senses the agricultural factors and recommends the favorable condition

of each relevant factor and finds the relevant model using the relevance vector analysis.

Keywords: Information Communication Technologies (ICT), Internet of Things (IoT), Precision

Agriculture (PA), Precision Agricultural Recommendation Model (PARM), Wireless Sensor

Networks (WSNs), Relevance Vector Analysis (RVA).

Tierärztliche Praxis

ISSN: 0303-6286

Vol 39, Issue 11, November - 2019

241

mailto:[email protected]:[email protected]

1. Introduction

Many studies say about 9.8 billion people will exist on earth by 2050, which in turn

increases the demand for food and lead in the malnourished nation. Hence this situation brings

the need to develop a ‘Precision Agriculture’ to meet this requirement and enhances the

agricultural procedure. The principle of this research is to design a new framework for increasing

the agricultural yield with less involvement of the farmers. Each farmer has a responsibility to

perform the sequence of actions in the field like sowing, weeding, watering, fertilizing and

finally harvesting. This process includes much manual labor intensive. Also, this includes many

decision-making processes to be made prior to more effective throughout the agriculture cycle.

Precision Agriculture or Smart Agriculture helps to address these issues by saving water, saving

fertilizer through effective usage and increasing the crop yield by continuous sensing the soil

moisture, humidity,temperature, and other supporting tasks. In many developed countries,

Precision agriculture which is completely IoT based agriculture is successfully running to

increase the yield of the crop S. Yao et al [9]. The two major challenges in precision agriculture

are to bring the awareness of utilizing the procedure of the IoT based framework into operation

among the farmers and the other is economical implementation. These challenges are balanced

by the effective utilization of sensors at low estimation.

Advances of Information Communication Technologies (ICT) are widely used in a

variety of fields such as business management, medicine recommendation, weather prediction

models, education system and communication system. But its contribution is not much used in

agriculture which determines the wealth of each nation. Most of the people living in rural areas

depend on agriculture and do not have facilities of technology in the agriculture sector to yield

high production. In our work, we anticipated a design framework for precision agriculture using

the advances of the Internet of Things and interpreting the findings to the farmer with the aid of

Relevance Vector Analysis (RVA) algorithm to provide the decision for cultivation. This system

helps the farmers in examining crop features and nature for giving require recommendations

about the components like temperature, water level, lighting, soil humidity and each aspect of

crop yield and finding proper composts to increase the yield in a simple justifiable common

dialect which certainly change conventional farming into precision or smart agriculture. In the

Internet of Things (IoT), the sensor plays a major role. The connection between these sensing

units is done either with wired or wireless networks.

In Wireless Sensor Network (WSN), numerous sensor nodes are interconnected with

limited processing capacity with more replicated data and this information are transmitted to the

receiver node and other sink nodes. This is treated as one of the downsides of the already

existing methods. In order to consider the accurate data with optimal network performance,

sharing and integrating the associated data between the sensor nodes and to consume the reliable

resource, a unique paradigm based on new advances of IoT is required. To fulfill this

requirement, an IoT based analytical model is proposed for the precision agriculture for the good

yield. Since WSN network is data-centric, data processing is a key problem in the WSN network

which in turn determines the performance of the wireless sensor network. Usually, each sensor

node is battery powered with limited energy. This research work focuses every process like data

sensing and collecting, data transmitting and data receiving, data storing and finally decision

making. In precision agriculture, the vast data are acquired and processed with the aid of WSN.

This idea of Precision Agriculture (PA) using Wireless Sensor Networks (WSNs) Lea-Cox et

al.[13], Green.O et al. [15] is very supportive for the farmers to increase the yield of the crops

along with applying modern technologies as a replacement for of customary farming practices.


ISSN: 0303-6286


242

1.1 IoT Action for Agricultural issues The main agricultural issues faced by the farmers are

a) Weather Condition The change in climate due to weather impact is a serious issue for agriculture. From the

observation made based on the climate, it is believed from the several experts of agriculture

sectors that the yield of agricultural production will decrease to 30% by 2050 due to the

adverse change of the climate. Climatic changes influence specifically every one of the

variables identified with agriculture and straightforwardly influences the quality and amount

of the product yield. A quick relief to this problem is applying the knowledge gained by the

information and communication technologies (ICT) Mizunuma et al.[14]. Thus Internet of

Things (IoT) can help us to allow, adapt and addresses the climate change.

b) Disease Detection and Diagnosis: The crop yield is highly affected by the several plant diseases and lack of proper pesticide

control mechanism. IoT enabled sensors to capture the image of affected plants and also

provides the proper prior decision for saving the life of the plant. Here image preprocessing

step helps the plant pathologists for the disease detection and proper diagnosis.

c) Usage of Fertilizer: The quantity of the fertilizer applied during the farming activity has the major

responsibility in saving the life and yield of the plant. The accurate decisions on the usage

of chemicals based on crop specification should be made by each farmer. The impact of

fertilizer usage not only kills the nutrient content of the plant product but also remain as a

major reason for many critical diseases.

d) Study on Soil Variations: The main success of agriculture is the soil which is treated as a major component in

farming. The soil moisture is sensed and monitored automatically under precision

agriculture. The modern devices for agriculture monitoring are available with monitoring

soil status such as gamma-radiometric soil sensors, a soil moisture sensor, EC sensors and

so on. The other sensors are used for monitoring the climatic condition and enumerating

the physiological status of the crops.

e) Water quality and quantity estimation: The estimation of the quantity of the water required for agriculture should be decided

earlier by the farmer. This is not a single factor depended; it depends on various factors

like climate, season, soil type, crop types, and growth stage of the crops. In general, during

cultivation, the crop loses water through transpiration and evaporation. Hence the sensors

are used to monitor the water level, the temperature of the water and water clarity.

f) Yield Production Readiness Analysis (YPRA): This analysis is based on the selling price of crops. The farmers make a clever

decision to sell the crops in peak demand once they know the information of the crop price

in advance. For this, a specific smartphone based sensors are designed to find out the ripping

condition of the fruits in well advance[13]. In some IoT based precision agribusiness,

advanced mobile phone camera is used to catch pictures of fruits under white and UV-A

light sources to decide ripeness levels for green fruits. Agriculturists could coordinate the

framework into their farms to isolate products of fruits of different ripeness levels into piles

before sending them to business sectors.


ISSN: 0303-6286


243

1.2 Benefits of Precision Agriculture:

Precision Agriculture Mohamed et al. [1] provides many benefits for promoting the

yield of agriculture. Some of the main issues related to farmers for agriculture welfare are

discussed below to resolve by direct integration.

a. Initial expectations and advantages promised un-fulfilled To begin with, every farmer ought to get more certainty to manage inconstancy and to

work in association with PA specialists. On the other hand, due to lack of support, they

become frustrated with the technology.

b. Complex technology with illogicality equipment The farmers should adapt to the technical issues of hardware and software. Also, the

farmers should aware of the illogically functioning components. The rate of adoption of PA

is relied upon to keep on rising in view of more prominent attention to the advantages of the

innovation.

c. Lack of products Due to lack of products, the PA integrates both the engineering and technology to the

successful farming. This idea integrates data collection, data processing, and other

monitoring process.

d. Increase in cost due to maintenance Due to the use of many networking things, the start-up and maintenance costs are so

high, which brings hesitation among farmers to introduce PA techniques.

2. Literature Survey

Agriculture is the most vital areas of human activity worldwide. This should be more

concentrated to balance population growth. Olmstead and Rhode [9] observed that the modern

agricultural system due to the commercialization, labor-intensive production and land resources

saving between 1870 and 1920s increases the agricultural manufacture double per land area. It

also the input resources used between the year 1920 and 1970 increases by about 30 % with the

increase in output of about 80%. Additionally noticed that yield increment was obviously not

only an expansion in the measure of data sources utilized, yet rather the innovation know-how

for effective farming information sources are used.

Recently, many research reasoned that the utilization of the biological developments,

collecting and sifting machines, mechanical innovation, and synthetic manures are the primary

calculates that expansion agricultural profitability per farmer three folds in the vicinity of 1970

and the 2000s Since fifteen years, the agriculturists began utilizing Information and

Communication Technologies(ICT) to systematize their money-related information and monitor

their business interchanges with outsiders and furthermore screen their harvests all the more

adequately. Now a day, in this Internet period, the data assumes an essential part of an

individual’s life. In this manner the horticulture is quickly turning into exceptional information

concentrated industry where farmers need to gather and assess an immense measure of data from

an alternate number of gadgets, for example, sensors, cultivating hardware, meteorological

sensors, and so forth to end up plainly more proficient underway and imparting proper data. The

endeavors cover all means in the generation chain involving the day by day agricultural tasks, the

value-based exercises for every single included partner and the help of data straightforwardness

in the food chain N.Wang et al [7].

Sorensen et al. [20] reported that farmers often experience an overload of information,

which create from different data sources and is represented in a variety of forms. Wang et


ISSN: 0303-6286


244

al.[19]applied the sensors to monitor temperature, soil moisture, and humidity and also

communicated the sensed information to the farmers by informing through the third parties like

meteorological stations. At that point, farmers joined every one of this information easily and

take exact choices F. Lan et al. [10] to deliver subjective items, enhance their salary and stick to

legislative directions and standards. A similar thought is additionally talked about by McCown et

al. [21], in this data gathered joined with the farmer’s interior arrangement of down to earth

knowing, learning and building a genuine intellectual framework.

Allen and Wolfert [16] proposed numerous proprietary solutions to help the farmers in

effectively monitoring their farms. Nikkila et al. [18] found more complex frameworks that track

topographical regions, climate designs and play out various propelled expectations. In later days,

Farm Management Information Systems (FMIS) concentrates on particular undertakings and

functional specification of the farms. Currently, these frameworks are gradually moving towards

the Internet time and use settled systems administration answers to enhancing agricultural

frameworks. In any case, it is broadly acknowledged that the Internet faces various deficiencies

particularly taking care of the huge quantities of organized gadgets either as the Internet of

Things or stakeholders. Yet at the same time, there is no institutionalized answer to empower

basic and strong interoperability among administrations and stakeholders. Henceforth it is trusted

that the Future Internet (FI) frameworks are relied upon to deal with these deficiencies.

C. Ayday et al.[17] identify two broad areas of application for precision agriculture based

on IoT which could be used to gather and analyze information for tracking the movement of

products through the supply chain or report on environmental conditions by monitoring soil

moisture or weather conditions. The IoT helps in automatic control by converting the collected

information into actions through a group of actuators. Also, it helps in optimizing processes,

resource consumption and to manage complex autonomous systems. Based on these

interconnections, the GSMA estimates that the amount of allied IoT devices will increase from 9

billion in 2012 to 24billion devices in 2020.

The application of sensor technology in the agricultural domain for solving the issues

related to the yield and for proper monitoring is proposed by M. R. M. Kassim et al. [1].

H.Sahota et al. [5] elucidated the sensor technology for the agricultural area by working on MAC

and Network layers. I. Mampentzidou et al. [2] provided the fundamental role of sensor

technology along with the essential components in the field of agriculture. Precision Agriculture

Monitor System (PAMS) for monitoring the agricultural work using sensor is proposed by S. Li

et al. [3].IFarm Framework system is suggested by Y. Jiber et al. [4] for managing the water

consumption for improving the production by enhancing the socio-economic factors. N. Wang et

al. [7] offered the outline for the latest designing of sensor technology for ecological nursing. M.

H. Anisi et al. [8] classified the sensor technology based on its performance metrics. Mafuta et

al. [6] emphasized the application of sensor infrastructure for monitoring and maintenance of

crops in an Orchard.

The main goal of this research paper is to suggest a functional model for the precision

agriculture named Precision Agricultural Recommendation Model (PARM) by utilizing IoT

abilities. Our objective is to construct an absolute monitoring system for improving the

agricultural functionalities with the innovations of the IoT. Using this innovative model the

farmer should be capable of managing all the shortcomings of today’s agriculture Y. Song et al.

[11], Pierce et al. [12].

The organization of the paper goes the way as the first section with an introduction about

precision agriculture, followed by the second section with discussing the shortcomings of an


ISSN: 0303-6286


245

earlier agricultural system with the modern system under literature survey. The third section

gives the proposed architecture named “Precision Agricultural Recommendation Model

(PARM)” with its observations. The fourth section includes some useful suggestions and

recommendations based on the test results for precision agriculture. Finally, the fifth section

winds up with the conclusion and future enhancement.

3. Proposed Architecture - Precision Agricultural Recommendation Model (Parm)

This research framework proposed the importance of the Internet of things (IoT) for

precision agriculture to increase the yield of cultivation. This hypothetical model works with

three phases in order to implement the model. Phase 1 includes the deployment of the sensor in

the agricultural field, Phase 2 with the processing and storing the sensed information of the

agricultural factors to the Base Station (BS), which is further analyzed in the Central Server

System (CSS). Finally, the data analysis is carried out by the Relevance Vector Machine (RVM)

to observe the framework operations to achieve accurate farming through precision agriculture.

The working of the Precision Agricultural Recommendation Model (PARM) is done in three

phases. Phase 1 is the construction of Agricultural Factor Monitoring Sensor (AFMS). Phase 2 is

an interconnection of Central Server System (CSS) with Agricultural Factor Monitoring Sensor

through Base Station (BS). Finally, the third Phase is model analysis by Relevance Vector

Machine (RVM) to promote cultivation based on the recommendation provided based on the

sensing agricultural factors. Figure 1 depicts the overall architecture of the Precision Agricultural

Recommendation Model (PARM).

Figure 1: Architecture for the Precision Agricultural Recommendation Model (PARM).

Central Server system (CSS)

Data Analysis by Relevance Vector Machine (RVM) to promote

cultivation

Precision

Agricultural

Recommendation

Model (PARM)


ISSN: 0303-6286


246

Phase 1: Components of Agricultural Factor Monitoring Sensor (AFMS)

The sensor is the basic receiving unit aids in monitoring the environmental factors and

output the sensed information. Sensors mainly play a vital role as the Environmental Information

Monitoring System (EIMS). The sensor hub includes various units such as Sensors, processor,

power supply and at last wireless transceiver unit, which forms the noteworthy part of the sensor

hub. Every sensor node plays out the gathering of operations to perform recognition,

accumulation, preparing and communicating the sensed data. The hardware organization of

sensor is shown in figure 2.

Figure 2: Hardware Organization of Typical Sensor

The operation of each unit as follows.

Sensing Unit: IT changes overall deliberate physical amounts, for example, temperature, dampness is changed over into a voltage flag and digitize it to

deliver computerized yield for preparing.

Processing Unit: The elements of the sensor nodes and deals with the correspondence conventions to do particular assignments are controlled by the

microcontroller of the handling unit.

Transceiver Unit: Transceiver unit plays out the Communication between the WSN node and the base station.

Power Unit: At last the power supply part that is the most urgent unit of sensor nodes provides required energy to these units.

The overall structure of the sensor module along with its communication layer is shown

in figure 3.


ISSN: 0303-6286


247

Figure 3.Structure of Sensor Nodes

In the agricultural field, the sensors are placed at a particular distance. Different types of

sensors such as soil moisture sensor, air temperature sensor, light intensity, humidity sensor, and

pH value sensor. In this research work, many sensors are used. Some of the sensors are utilized

to measure soil temperature, soil moisture, soil humidity and other factors promoting the high

yield.

In general, the yield of the crop is determined using the following equation (1). For the crop

yield determination, the grain flow rate and the area covered are considered for the yield

calculation,

Mass Volume

AreaArea Yield =

(1)

Each agricultural parameter is measured by the sensor is used as the input of the control

unit. For instance, the moisture sensor measures the moisture level either by the Volumetric

Water Content Unit (VWC) or gravimetric water content Unit (GWC). In the case of

Volumetric Water Content Unit (VWC), the sensor measures the moisture level as a numerical

measure and is measured as the ratio of water volume to the soil volume. But in the case of

gravimetric water content (GWC), the measurement is considered as a weight rather than the

volume. Some default and suitable levels of the agricultural factors for the crop growth which is

determined by the agronomist is shown below in Table 1.


ISSN: 0303-6286


248

Table 1.Effect of climatic factors on crops

The cultivation factor for the Agricultural Factor Monitoring Sensor (AFMS)system

includes certain sensors that could be used for measuring the humidity, pH of the soil, moisture

level and temperature. In general, each sensor is available as a single unit in the market. For

example, if we take a sensor MTS400, then it could contain the relative activity of the sensors

that are mentioned above. Sometimes data acquisition boards such as MDA300 could also be

attached that could sense the moisture level of the soil. These boards are companionable with the

processor unit which performs processing and storing the sensed data. A diagram that represents

the sensor board with its processor and data acquisition board in an agricultural farm with the

appropriate housing is shown below in the figure 4.

Agricultural

Stages

Air Temperature Soil Temperature Soil Moisture Humidity

Sprouting 26-33 degree Celsius

(Best). At least 18

degree Celsius

(Minimum)

23-28 degree Celsius

(Optimum)

Provided by water

absorbed

-

Tillering Cool nights Minimum when the soil

is warm

Assisted by

adequate moisture

in the soil

-

Growth 30-33 degree Celsius

(Best) and poor when

< 20 degree Celsius

23-29 degree Celsius

(Optimum) and poor

when < 21 degree

Celsius

Essential of

adequate moisture

Better

Flowering Preferred warm nights

or with 18 degree

Celsius

Utmost in warm soil. Best in moist soil

and stopped at

drought.

Better

Ripening Preferred cold nights

or Optimum < 15

degree Celsius

Low Temperature is

best

provoked at a

minimum moisture

Preferred

dry

climate

Over-

Ripening

provoked at hot

season

Favored by high

temperature

Favored by water

availability during

the dry period.

-


ISSN: 0303-6286


249

Figure 4.Deployment of various sensors in the agricultural field

3.1 Determination of requirement of sensor nodes

To estimate the required amount of sensor for the given agricultural ground as

per the planned model is determined by equation 2.

1 1 2 ( -1) (( ) 1 ( ) 1 2d d dN N f N f N fs w w wf fw w

(2)

Where Ns represents the number of required sensor nodes. dN represents a number of divisions

in the agricultural field and fw represents the field width in the agricultural land. The sensor

nodes are distributed according to the field width.

3.2 Deployment of sensor nodes

The usual method of planting plants in each division of field depends on the size of the

agricultural field. This way of division system is applied to distribute the sensor nodes

consistently and also used for node conversion and identification.

Here we assume that the land is divided into equal tubs of one unit area. Each tub will have

two sensor nodes that are dispersed on it with the estimated partition of a fixed meter (here eight

meters). The deployment of each node is determined by its specific location which wills help to

find the fault of the sensor node. This procedure is suitable for the fault tolerance and checking

the connectivity of the nodes based on their separation. The preferred and suitable distance

between each sensor nodes is 10 m for sensing the microclimate throughout the entire growing

season. The sample deployment of sensors in the agricultural field is shown below in figure 5.


ISSN: 0303-6286


250

Figure 5. Deployment of the sensor in the field

The optimal values for the climatic factors such as air temperature, soil temperature, soil

moisture, and soil humidity are listed above in Table 1., this favors the primitive cultivation for

precision agriculture. Based on the sensed value provided by each sensor to the Base station

(BS), the Central Server System (CSS) will have the backup of all the data and the data analytics

is carried out to represent the optimal condition for the precision agriculture.

As the data analytics, Relevance Vector Machine (RVM), a well known statistical

analysis algorithm is utilized to predict the relevance factors based on the sensor value for the

high yield. This algorithm depends on the least number of relevance vectors (i.e., training

samples).In this precision agriculture model, the Relevance Vector Machine depends on the

training samples and their associated hyperparameters which leads to a sparse solution.

3.3. Data Analysis by Relevance Vector Machine (RVM)

The Relevance Vector Machine (RVM) is implemented in MATLAB with the K-fold cross-

validation (here k=5 is considered). Generally, this particular validating method is an efficient

method to evaluate the training and test samples.

In our model, the Relevance Vector Machine (RVM) is applied to find the position of

the land where all the climatic factors are in good condition and preferable to promote the

cultivation. Based on the data stored in the Central Server System(CSS),the data analysis is

carried out by the Relevance Vector Machine (RVM) to classify the land or find the position of

the land to reveal the factors which are not in satisfied condition and whether to use or not use

the that land for the cultivation. The model was created with an informed choice process trying

to clarify the information in the least difficult way imaginable. Potential sources of input were

analyzed to see which were most applicable to the objective capacity and in this manner abstain

from debasing the execution of a learning calculation because of the nearness of unessential

information factors. In every cycle, the contribution of input with the least efficiency was

dispensed. Also, the performance of this method depends on the crop growth and its yield. This

is finally observed and analyzed by crop growth Image and was captured by using a JEPG

module and stored for future evidence (shown in Algorithm 3). Figure 6 shows the system flow

of the proposed architecture.


ISSN: 0303-6286


251

Figure 6. Agricultural Factor Monitoring Sensor (AFMS) system

The Root Mean Squared Error (RMSE) as well as the Nash–Sutcliffe efficiency (NSE), both are

utilized to assess the precision agricultural model. The bigger the estimation of NSE, the little

the estimation of RMSE the greater value of precision and accuracy of the model aid to estimate

Agricultural Parameter Attentiveness Level (APAL). The RMSE and NSE are registered as

appeared in Equations (3) and (4) separately:

Are they

better?

Start

Soil Temperature

Sensor Soil Humidity

Sensor

Soil Moisture Sensor

Central Server

System (CSS)

Check and compare the observed Value with the

estimated value (threshold value)

Search for the better

conditions

Data Analysis by Relevance

Vector Machine (RVM)

Yes

No

Decision making for the Precision

Agriculture Recommendation

Start


ISSN: 0303-6286


252

N2

p p

p=1

ˆ( y -y )

Root Mean Square Error (RMSE) =N

(3)

2N ˆ(y - y)pt=1

=Nash – Sutcliffe effi 1 -2N (y - y

ciency E)pt=1

(4)

Where ˆ py is the forecasted agricultural parameter, py is the calculated agricultural parameter,y is

the mean of observed agricultural parameter concentration, ŷ is the mean of estimated observed

agricultural parameter concentration and N is the total number of the observations.

For Precision agriculture, every Agricultural Parameter Attentiveness Level

(APAL) is estimated with the evaluation of comparing the Root Mean Squared Error (RMSE) as

in equation (3) and the Nash–Sutcliffe efficiency (E)as in equation (4). The parameter which

holds the smaller RMSE and larger E is considered as sufficient parameter and then chosen those

condition for the cultivation. This mechanism brings the evidence-based agricultural model for

increasing the yield of agriculture. The procedure for sensing the agricultural factors is given in

algorithm 1 and algorithm 2.

Algorithm 1: Sensors that monitor the humidity and temperature

Step 1: The soil temperature and humidity are sensed by sensor initially and commit the

information at a baud rate of 9600.

Step 2: Read the value obtained from the allocated analog pin of respective humidity and

temperature.

Step 3: Print the sensed temperature in degree Celsius and Humidity in percentage (%) on the

serial monitor

Step 4: Even at the delay of 2000ms continue to read data.

Step 5: Terminate and DE-initialize the sensor.

Algorithm 2: Sensing of Soil Moisture

Step 1: To direct data at a Baud Rate of 9600 sensors can be set to sense the soil moisture.

Step 2: From the assigned analog pin start reading the soil moisture level.

Step 3: if 1000 < moisture level

Print "Low"

Else if moisture level500

Print "Medium"

Else

Print "High"

Step 4: Prolong to read even at the delay of 2000ms

Step 5: Terminate and De-initialize the sensor


ISSN: 0303-6286


253

Algorithm 3: Storing the captured JPEG image of the crop growth in the SD card

Step 1: clear and initialize the buffer of the SD card. On the Arduino, the chip select pin is

assigned.

Step 2: Thus for assigning the byte stream the camera has to be initialized.

Step 3: The function SendCmdpic() is run that captures the image with the time-delay of 300ms.

Camera buffer takes the Picture and stores it into the file using SendReadDataCmd() command.

Proper delay time should be maintained while reading and writing the data in such a way that it

should not affect the intermediate buffer.

For morphological analysis, the above 1-3 algorithms are used to confine the image. The

observation of crop growth is carried out under four following conditions.

The four observing conditions are

Conditions:

1. Observing the crop under normal healthy conditions. 2. Observing the crop under the excessive conditions of both sunlight and moist conditions.

3. Observing the crop beneath minimum to a reasonable temperature of the sunlight with modest to the absence of water condition.

4. Observing the crop by treating the crop under normal environmental conditions by providing an excess of fertilizers like N, P and K.

4. Test and Results

Figure 7 shows the data for humidity and temperature during the experimental period.

There are high temperature and low humidity at the day time. Moreover, the evaporation of

water also takes place at a high rate during day time. This will result in the lowering of moisture

level at the day time as to the nighttime to attain threshold value.


ISSN: 0303-6286


254

Figure 7. Trail period with the Temperature and Humidity conditions

Test samples of 100 plants were picked arbitrarily for this experiment. The humidity

sensors were utilized to catch dampness information at a consistent interval of 60 minutes.

This particular irrigation project was carried out at a tropical climate at a time around 8

am,10 am, 12noon, 2 pm and 4 pm in a day. Irrigation was carried out at a duration of 5 minutes.

Cultivation with PARM model keeps the moisture condition above 35VWC (threshold

value). The irrigation pump will be set into action once the moisture level goes beyond the

threshold value. The utilization of water and fertilizer for irrigation purposes is shown in the

above two cases. It is necessary to maintain the water level for the plants to be healthy. The first

irrigation was carried out at Feburary, 12 pm that is shown in the figure above. Although the

VWC value obtained was greater than the threshold value this process was maintained.

The quantity of water and the quantity of the fertilizer is shown in the table given below.

From the table, it could be seen that the amount of water utilized by the scheduled mode of

irrigation varies at a rate of 2500ml, while the automatic mode of irrigation consumes only 100

ml of water. This has been estimated in a day.

Table 3 Consumption of water and fertilizer under Precision Agricultural Recommendation Model (PARM)

Mode of Cultivation Total Cultivation /day Water consumed /Cultivation

Total Water

Consumption

(ml)

Scheduled mode 5 500 2500

Automatic mode 5 200 1000


ISSN: 0303-6286


255

5. Conclusion and Future Enhancement

IoT is getting very fashionable and admired by connecting lots of devices to the internet

and easy accessing and analyzing gathered data. Agriculture as an open plant has a big potential

to use sensors and other devices to be connected to the internet since it is always very difficult to

get data for making decisions and tracking of products and machinery performance data. For IoT

applications to be integrated into agriculture even short or long-range communication technology

should be chosen properly to get data from long distances, save sensor energy and most

important for making it in the cheapest way. In these circumstances, “Precision Agricultural

Recommendation Model (PARM)” comes along as a new technology that enables IoT to be

integrated into agriculture, minimizes obstacles in data transmission and enhancing the precision

agriculture. As a future enhancement, this model is reframed with the aid of Gustafson-Kessel

(GK) clustering algorithm to save the resource and energy cost and to provide the decision for

cultivation

Referances:

[1] Mohamed Rawidean Mohd Kassim, Ibrahim Mat, Ahmad Nizar Harun,Wireless Sensor

Network in Precision Agriculture Application, IEEE, pp. (978-1-4799-4383-

8/14/©2014 IEEE).

[2] Ioanna Mampentzidou, Eirini Karapistoli, Anastasios A. Economides,Basic Guidelines

for Deploying Wireless Sensor Networks in Agriculture, (978-1-4673-2017-7/12/©2012

IEEE).

[3] Shining Li, Jin Cui, Zhigang Li, Wireless Sensor Network for Precise Agriculture

Monitoring, (978-0-7695- 4353-6/11 ©2011 IEEE) (DOI 10.1109/ICICTA.2011.87).

[4] Yassine Jiber, Hamid Harroud Ahmed Karmouch,Precision Agriculture Monitoring

Framework Based on WSN, (978-1-4577-9538-2/11©2011 IEEE).

[5] Herman Sahota, Ratnesh Kumar, Ahmed Kamal, Jing Huang, An Energy-efficient

Wireless Sensor Network for Precision Agriculture_, (978-1-4244-7755-5/10/©2010

IEEE).

[6] Million Mafuta, Marco Zennaro, Antoine Bagula, Graham Ault,1, Harry Gombachika,

and Timothy Chadza, “The Application of Wireless Sensor Networks in Management of

Orchard_ D. Li and C. Zhao (Eds.): CCTA 2009, pp. 519–522, 2010. © Springer-

Verlag Berlin Heidelberg 2010.

[7] N.Wang, N. Zhang, and M.Wang, “Wireless sensors in agriculture and food industry

recent development and future perspective,” Computers and electronics in agriculture,

vol. 50, no. 1, pp. 1–14, 2006.

[8] M. H. Anisi, G. A. Salaam and A. H. Abdullah, A Survey of Wireless Sensor Network

Approaches and their Energy Consumption for Monitoring Farm Fields in Precision

Agriculture7, Precision Agric, Springer, 2015

[9] S. Yao, C. Feng, Y. He, and S. Zhu. “Application of IOT in agriculture,” Journal of

Agricultural Mechanization Research, vol. 07, pp. 190-192, 2011.

[10] F. Lan. “Green house precise management system based on production rules,” Journal

of Agricultural Mechanization Research, no. 2, pp. 80-83, 2012.

[11] Y. Song, X. Qiao, J. Wang, W. Zheng, and X. Zhang. “Application actuality and trend


ISSN: 0303-6286


256

of RFID in agriculture, Journal of Agricultural Mechanization Research, vol. 10, pp. 1-

5, 2010.

[12] Pierce, F.J., Elliott, T.V.: Regional and on-farm wireless sensor networks for

agricultural systems in Eastern Washington. Computers and Electronics in Agriculture

61(1), 32–43 (2008)

[13] Lea-Cox, J.D., Ristvey, A.G., Arguedas, F.R., et al.: Wireless sensor networks for

realtime management of irrigation and nutrient applications in the greenhouse and

nursery industry. Hortscience 43(4), 1103 (2009)

[14] Mizunuma, M., Katoh, T., Hata, S.: Applying IT to Farm Fields A Wireless LAN. NTT

Techinical Review 1(2), 6–60 (2003)

[15] Green, O., Nadimi, E.S., Blanes-Vidal, V., et al.: Monitoring and modeling temperature

variations inside silage stacks using novel wireless sensor networks. Computers and

Electronics in Agriculture 69(2), 149–157 (2009)

[16] Allen, J., Wolfert, J. Farming for the Future: towards better information based

decision-making and communication - Phase I: Australasian stocktake of farm

management tools used by farmers and rural professionals. New Zealand Centre of

Excellence in Farm Business Management, Palmerston North.

http://edepot.wur.nl/194811. 2011.

[17] C. Ayday and S. Safak, “Application of wireless sensor networks with GIS on the soil

moisture distribution mapping,” in Symposium GIS Ostrava, 2009.

[18] R. Nikkila, I. Seilonen, K. Koskinen, Software architecture for farm management

information systems in precision agriculture, Comput. Electron. Agric., 70 (2010) 328-

336.

[19] J Wang, J Huang, S Rozelle, Q Huang, A Blanke, Agriculture and groundwater

development in northern China: trends, institutional responses, and policy options,

Water Policy, 9 (S1), 61-74.

[20] C.G.Sorensen , S.Fountas, E.Nashf, L.Pesonend, D.Bochtisa S.M.Pedersene B.Bassoc

S.B.Blackmoreg, Conceptual model of a future farm management information system,

Computers and Electronics in Agriculture Volume 72, Issue 1, June 2010, Pages 37-47

[21] R.L. McCownc P.S.CarberryaN.P.DalglieshaM.A.FoalebZ.Hochmanc, Farmers use

intuition to reinvent analytic decision support for managing seasonal climatic

variability, Agricultural Systems,Volume 106, Issue 1, February 2012, Pages 33-45.

AUTHORS PROFILE

Mr P.Suresh, M.E, is an Assistant Professor(Sl.G) in the Department of Computer Science and Engineering at KPR

Institute of Engineering and Technology.Prior to this he was associated with VelTech MultiTech Dr.Rangarajan

Dr.Sakunthala Engineering College, Chennai. He obtained his B.Tech(IT) from VelTech Engineering College

(Anna University, Chennai) and M.E (CSE) from VelTechMultitech Engineering college (Anna University,

Chennai). He is pursuing his Ph.D in Vel Tech Rangarajan Dr. Sagunthala R&D Institute of Science and

Technology,Chennai. He has been in the teaching profession for the past 10 years and has handled both UG and PG

programmes. His area of research is Internet of Things (IoT).He has published 3 books, 10 research articles in

International Journals and 10 papers presented in International and National Conferences. He has attended various

Training Programmes, Workshops& FDPs sponsored by AICTE related to his area of interest. Recently, the

students won Project Innovation Awards for an IOT project under his guidance in various contest and applied the

same for patent.He is a life time member of IAENG.


ISSN: 0303-6286


257

javascript:void(0)javascript:void(0)http://www.sciencedirect.com/science/article/pii/S0168169910000396?via%3Dihub#!http://www.sciencedirect.com/science/article/pii/S0168169910000396?via%3Dihub#!http://www.sciencedirect.com/science/article/pii/S0168169910000396?via%3Dihub#!http://www.sciencedirect.com/science/article/pii/S0168169910000396?via%3Dihub#!http://www.sciencedirect.com/science/article/pii/S0168169910000396?via%3Dihub#!http://www.sciencedirect.com/science/article/pii/S0168169910000396?via%3Dihub#!http://www.sciencedirect.com/science/journal/01681699http://www.sciencedirect.com/science/journal/01681699/72/1http://www.sciencedirect.com/science/article/pii/S0308521X11001557#!http://www.sciencedirect.com/science/article/pii/S0308521X11001557#!http://www.sciencedirect.com/science/article/pii/S0308521X11001557#!http://www.sciencedirect.com/science/article/pii/S0308521X11001557#!http://www.sciencedirect.com/science/article/pii/S0308521X11001557#!http://www.sciencedirect.com/science/journal/0308521Xhttp://www.sciencedirect.com/science/journal/0308521X/106/1

Dr. Koteeswaran Seerangan, currently working as an Associate Professor in the Department of Computer Science

and Engineering at Vel Tech Rangarajan Dr. Sagunthala R&D Institute of Science and Technology, Chennai-62,

Tamilnadu, India. He has authored and co-authored several papers in various reputed journals and conference

proceedings. He is a reviewer for more than a dozens of Conference Proceedings and Journals. His research

interests include Data Mining, Internet of Things, Big Data and Analytics. He is a Member of ACM, Member of

IET, Senior Member of IEEE and Life Member of ISTE.

Ms.R.H.Aswathy, B.E, M.E, is Assistant Professor (Sr.G) in Department of Computer Science and Engineering,

since June 2017. She obtained her B.E (CSE) from Narayanaguru Engineering College (Anna University, Chennai)

and M.E (CSE) from Karpaga vinayaka college of Engineering and Technology (Anna University, Chennai). She

has been in the teaching profession for the past 6 years and has handled both UG and PG programmes. Her area of

research includes IoT and Artificial Intelligence. She received faculty partnership model award-Bronze level twice

from Infosys in the year 2016 and 2017.She has published 3 books. She has published 7 research papers in

International Journals and 7 papers in International and National Conferences. She has attended many workshops &

FDPs sponsored by AICTE related to her area of interest. She is a life time member of IAENG.

Dr. N. Malarvizhi, currently working as Professor & Head in the Department of Computer Science and

Engineering at Vel Tech Rangarajan Dr. Sagunthala R&D Institute of Science and Technology, Chennai-62,

Tamilnadu, India. She is having more than 15 years of teaching experience. She has written a book titled

“Computer Architecture and Organization”, Eswar Press, The Science and Technology Book Publisher, Chennai.

She serves as a reviewer for many reputed journals. She has published numerous papers in International

Conferences and Journals. Her area of interest includes Parallel and Distributed Computing, Grid Computing,

Cloud Computing, Big Data Analytics, Internet of Things, Computer Architecture and Operating Systems. She is a

life member of Computer Society of India (CSI), IARCS and IAENG. She is a Senior Member of IEEE and IEEE

Women in Engineering (WIE).

Mr Balamurugan C, M.E, is an Assistant Professor in the Department of Computer Science and Engineering at

V V collegeof Engineering .Prior to this he was associated with Vel Tech Rangarajan Dr.Sagunthala R&D Institute

of Science and Technology, Avadi , Chennai. He obtained his B.E(CSE) from Dr.Sivanthi Aditanar College of

Engineering College (Anna University, Chennai) and M.E (CSE) from VelTechMultitech Engineering college

(Anna University, Chennai). He has been in the teaching profession for the past 9 years and has handled both UG

and PG programmes. His area of research is Internet of Things (IoT).He has published 6 research articles in

International Journals and 3 papers presented in National Conferences. He has attended various Training

Programmes, Workshops& FDPs related to his area of interest.


ISSN: 0303-6286


258
https://veltech.edu.in/https://veltech.edu.in/

an iot based innovative irrigation model for smart and ...tierarztliche.com › gallery ›...

Documents