© 2024, Centre for Advanced Research in Agricultural Sciences
Research Journal of Agricultural Sciences
Volume 15; Issue 03 (May–Jun 2024); pp 672–673
Artificial Intelligence in Irrigation
Sandeep Bhardwaj*
1
and Rupali Sharma
2
1
Department of Basic Engineering, COAE&T, CCS Haryana Agricultural University, Hisar - 125 004, Haryana, India
2
Department of Horticulture, CCS Haryana Agricultural University, Hisar - 125 004, Haryana, India
Received: 05 Feb 2024; Revised accepted: 20 Apr 2024
Key words: Irrigation, Irrigation schedules, Optimization in agriculture, Water usage
Irrigation practices are primitive in nature in India as
compared to developed countries. Irrigation extensively
influences the crop yield and farmer’s income. Agriculture is an
unorganized sector still away from modern technology reach in
India. Irrigation techniques and marketing businesses are
governed by brokers and middle men in India and this lead to
unnecessary wastage of water as explained in [1]. However,
provides the Automatic Irrigation System based on Artificial
Intelligence and Internet of Things, which can autonomously
irrigate fields with use of soil moisture data. A prediction
algorithm applied over past weather data for rainfall [2].
Refuses to provide any code as a demo but explains with its
flow chart their system on irrigation with selective crop field
data in real-time soil moisture conditions [3].
Agricultural Automation was increasingly beneficial
through the use of [4] the Artificial Intelligence and Machine
Learning. Arduino and Raspberry devices may be embedded
with moisture and temperature sensors along with the help of
Machine Learning algorithms [5]. Data storage and
management of sensors is in the online cloud [6]. Highlighted
need of Automatic irrigation in the evapo-transpiration process
and the crop prediction. Lastly, emphasis on the Artificial
Intelligence and embedded systems in the irrigation of
agriculture [7].
Irrigation for a wheat crop requires extensive irrigation
planning from flood irrigation methods to drip irrigation
methods in sub-tropical climates. Classification of wheat
irrigation systems using Artificial Neural Network as given in
[8] uses a computer vision system. Classify wheat irrigation of
the crop species Triticum aestivum and Triticum durum
according to their visual irrigation characteristics using an
artificial neural network of the MLP type [9]. The images are
obtained by a camera at an angle perpendicular to the plot.
These images converted to grayscale, binarized using the Otsu
method and segmented by the thresholding operation. The
characteristics of plot size, fertigation techniques and
availability of water are feeded for each row of the plot, with
the purpose of serving as input to the classification method in
Artificial Intelligence. Following visual characteristics of the
irrigation were selected: length, length and width ratio, green,
and blue, green ratio, homogeneity, and entropy. The last two,
concerning texture, are obtained using the GLCM method. The
ANN (Artificial Neural Network) based on MLP with three
layers classified irrigation in bread wheat and durum wheat
irrigation systems as explained in [3]. Option of selective
irrigation systems in the wheat crop may be possible as shown
in [3].
Survey techniques are important in modern irrigation
systems. Traditional irrigation technologies rely on manual
approach and are time consuming and labor intensive.
Unmanned Aerial Vehicles (UAVs) equipped with various
sensors are good for the surveying procedure, decrease data
collection time, and reduce cost as explained in [10]. This will
apply accurately and rapidly in the field and helps in analysis
and visualization of collected data from UAVs and similar kind
of small airplanes, satellites, ground platforms. This UAV data
collected in the cloud and artificial intelligence (AI) based
application (named Agroview) which interact with user-
friendly application helps in detection, count and geo-location
of the plot and plot size (locations in terms of GPS data and
local landmarks). It further measures the plot irrigation
efficiency. And helps in developing the plant health (or stress)
maps to measure the irrigation efficiency. This also provides the
specified need of irrigation for a particular row or for a whole
plot as explained in [11].
SUMMARY
Artificial intelligence can also be used to predict weather
patterns and adjust irrigation schedules accordingly, reducing
the need for manual adjustments based on weather forecasts.
This can lead to significant water savings and improved crop
yields. In terms of optimizing crop selection and planning, AI
can analyze historical data and environmental factors to
recommend the most suitable crops for a specific region, taking
into account factors such as soil type, climate, and water
availability. This can help farmers make more informed
decisions about what crops to plant and where, leading to higher
yields and reduced water usage. Overall, the use of AI in
irrigation has the potential to significantly reduce water usage,
improve crop yields, and minimize environmental impacts
associated with agriculture. As technology continues to
advance, we can expect to see even more innovative
applications of AI in water management and agriculture
specifically on weather forecasts. This can lead to significant
water savings and improved crop yields. In terms of optimizing
crop selection and planning, AI can analyze historical data and
environmental factors to recommend the most suitable crops for
CARAS
Short Communication
*Correspondence to: Sandeep Bhardwaj, E-mail: bhasandeep@gmail.com; Tel: +91 8901375772
Citation: Bhardwaj S, Sharma R. 2024. Artificial intelligence in irrigation. Res. Jr. Agril. Sci. 15(3): 672-673.
ISSN: 0976-1675 (P)
ISSN: 2249-4538 (E)