https://doi.org/10.1140/epjs/s11734-021-00333-4
Regular Article
Fractal analysis and machine-learned decision system for precision and smart farming
1
University School of Basic and Applied Sciences, Non-Linear Dynamics-Research-Lab, Guru Gobind Singh Indraprastha University, Dwarka, New Delhi, India
2
Maharaja Agrasen Institute of Technology, Guru Gobind Singh Indraprastha University, Delhi, India
3
Department of Mechanical Engineering, College of Engineering, Qassim University, 51452, Buraidah, Al Qassim, Saudi Arabia
Received:
15
July
2021
Accepted:
30
October
2021
Published online:
6
December
2021
Agriculture is the main source of income for most of the people from thousands of years but still every year farmers suffer loss of crop and money, due to misinterpretation of soil or climatic conditions. In the recent years, researchers have worked to improve this state and agriculture for production of crop by analyzing soil or climate conditions. In this paper, we proposed the methods as fractal analysis and machine-learned decision system for smart and precision farming. The long-term behavior of the different parameter on which production of crop depends is analyzed using fractal analysis which is helpful for crop maintenance and also helpful in preparing the framework for the government and farmers in advance to know the total automating of the supplements, water framework, water system controls for accuracy of cultivating in the fields. Using Hurst exponent and Fractal analysis, it is observed that all the seven parameters affecting the crops follow anti-persistent behavior which shows the drawn out exchanging among high and low qualities with a definite pattern. Machine-learned system including artificial neural network, kNN Classifier, XG Boost, Random Forest classifier conclude that different decision systems show the accuracy for different crops with parameters from 95 to 99%. Random Forest classifier gave more accuracy among all classifier for testing and providing support for crop management systems. It is concluded that the proposed technique using machine-learned classifier is giving more accuracy for precise and smart farming with good crop management and helpful for the government to make the decision and formulate policies for the stack of farmers, consumers and for the development of the nation as farmers are backbone of any country.
© The Author(s), under exclusive licence to EDP Sciences, Springer-Verlag GmbH Germany, part of Springer Nature 2021
