AI-Based Model to Assess Land Suitability for Crop Production Improvement
Abstract
The world population is expected to grow by another two billion in 2050, according to the
survey taken by the Food and Agriculture Organization, while the land used or suitable for
growing crops is likely to grow only by 5%. Therefore, smart and efficient farming
techniques are necessary to improve agriculture productivity. Agriculture land suitability
assessment is one of the essential tools for agriculture development. The rapid development
of wireless sensor networks has triggered the design of low-cost and small sensor devices
with the Internet of Things (IoT) empowered as a feasible tool for automating and decision-
making in the domain of agriculture. The proposed system is an implementation of sensor
networks integrating with Artificial Intelligence systems such as neural networks and Multi-
Layer Perceptron named as Artificial Intelligence Agriculture Recommendation
Model(AIARM) for the assessment of agriculture land suitability. Rather than binary
classification, the proposed system classifies the agriculture land into four decision classes,
namely more suitable, suitable, moderately suitable, and unsuitable to guide the farmers’
[Link] performance of the MLP-based multiclass classification model will be
compared with neural networks to deliver better results.
Problem Statement
• To develop a system which is an amalgamation of WSN and Artificial Intelligence
(neural networks and Multi-Layer Perceptron (MLP)) to appraise agriculture land
suitability.
• This model will help the agriculturists to assess the arable land in terms of four decisive
classes, namely more suitable, suitable, moderately suitable, and unsuitable.
• The appraisement is made formulated based on the input congregated from the various
sensor devices, which are used for training the model using MLP with four hidden layers
for multiclass classification of land felicitousness to ameliorate the crop productiveness
for sustainable agriculture development.
Literature Survey
Title : Integrative Use of IoT and Deep Learning for Agricultural
Applications
Source Methodology Description Conclusions
Authors:Disha Garg; [Link] is used for Input data is A comparison
Samiya Khan; Mansaf Hydroponic system collected from between two
Alam. development(multiple iot sensors like algorithms on
Publisher:Springe r input parameters) [Link] agriculture data to
Nature Switzerland [Link] Learning sensor, make predictions for
Year:2020 Framework to 2. Soil agricultural
perform deep CNN moisture sensor, applications.
training [Link] DNN=88% accuracy
sensor and Deep CNN=96.3%
4. Co2 sensor.
Title : A comprehensive review on automation in agriculture using artificial
intelligence.
Source Methodology Description Conclusions
Author:Kirtan [Link] different Input data is collected This system
Jha ; Aalap automation practices from iot sensors like presented an idea to
Doshi ; Poojan like IOT, Wireless [Link] sensor, develop a system
Patel ; Manan Communications, [Link] moisture sensor, with IOT and ML
Shahd Machine learning and [Link] ,to automate the
Publisher:KeAi Artificial Intelligence, sensor and traditional practices
Year: 2019 Deep learning. in agriculture
[Link] 4.Co2sensor.
techniques like Fuzzy [Link]
logic is used for land sensor(MQ4,MQ7 )
suitability prediction.
[Link] for Wireless
networks.
[Link] learning
Algorithms like ANN
to identify the soil
moisture
Title : IoT based Smart System to Support Agricultural Parameters
Source Methodology Description Conclusions
Authors:AbhijitPathak Cuckoo Search Input data is collected Data received in the
a, Algorithm has from iot sensors like ThingSpeak
MohammadAmazUddin, been developed, temperature, implemented by
Md. JainalAbedina, Karl allowing the turbidity, Cuckoo Search
Anderssonb,RashedMus allocation of water pH., Algorithm,
t afac, Mohammad for farming under moisture allowing the
Shahadat Hossainc any conditions. selection of
Publisher:Elsevier appropriate crops
Year:2019 for particular soil.
Title : Smart Sustainable Agriculture (SSA) solution underpinned by Internet of
Things (IoT) and Artificial Intelligence (AI).
Source Methodology Description Conclusions
Author: A holistic IoT/AI Input data is AI/IoT platform
[Link] idi platform to cover all collected from iot for SSA, to
Publisher:arXiv areas within a SSA sensors like resolve issues
Year:2019 environment(Smart [Link] resulting from the
Sustainable Agriculture ) sensor, fragmentary
to perform tasks like [Link] moisture nature of the
govern data sensor, agricultural
flow,integration of [Link] process.
components,sustainable sensor and
storage,etc. 4.Co2 sensor.
Title : Title : Internet of Things for smart precision agriculture and farming in
rural areas.
Source Methodology Description Conclusions
Author:Ahmed, [Link] WiLD Input data is collected The proposed fog
A.N.; de network and fog from iot sensors like computing solution
Hussain, I.D computing in the existing [Link] sensor, takes action with
Publisher:IEEE WSN-based solutions for [Link] moisture sensor, lesser delay and
Year:2018 covering longer range [Link] sensor saves bandwidth in
with lesser delay. and the network.
[Link]-layer-based MAC 4.Co2 sensor.
and routing solution is
used to reduce delay.
Title : IoT solutions for precision farming and food manufacturing
Source Methodology Description Conclusions
Author:Rob Ai algorithms like Input data is The parameter of malt
Dolci. Bayesian network collected from iot before and after the
analysis and multi sensors like development. were
Publisher:IEEE -variant analysis are [Link] sensor, shown on protein and
Year:2017 implemented to check [Link] moisture starch values to
how CO2, sensor, understand germination.
Temperature, [Link]
Humidity and PH sensor and The end result is malt
vary with higher content of
4.Co2 sensor. starch and proteins,
which in turn means
higher alcohol.
Architecture diagram
Architecture diagram
References
[1]. Ahmed, A.N.; de Hussain, I.D. Internet of Things (IoT) for smart precision agriculture and farming in
rural areas. IEEE Internet Things J. 2018
[2]. [Link], Smart Sustainable Agriculture (SSA) solution underpinned by Internet of Things (IoT)
and Artificial Intelligence (AI). arXiv 2019
[3]. Rob [Link] solutions for precision farming and food manufacturing Artificial Intelligence
applications in Digital [Link] 2017
[4]. Disha Garg;Samiya Khan;Mansaf [Link] Use of IoT and Deep Learning for Agricultural
[Link] Nature Switzerland AG 2020.
[5]. Kirtan Jha ; Aalap Doshi ; Poojan Patel ; Manan Shahd. A comprehensive review on automation in
agriculture using artificial [Link] 2019 6. AbhijitPathaka, Mohammad AmazUddina, Md.
Jainal Abedina, Karl Anderssonb, Rashed Mustafac, Mohammad Shahadat [Link] based Smart
System to Support Agricultural [Link] 2019
[6]. AbhijitPathaka, Mohammad AmazUddina, Md. Jainal Abedina, Karl Anderssonb, Rashed Mustafac,
Mohammad Shahadat [Link] based Smart System to Support Agricultural [Link]
2019
