Smart Crop Prediction using IoT and Machine Learning

Abstract

Agriculture is a crucial profitable motorist. It\'s a key to healthy biosphere. People depend on a wide range of agrarian products in nearly all aspects of life. growers need to manage with climate change, and meet rising demands for further food of advanced food quality. In order to escalate the yield and growth of crops, the planter needs to be apprehensive of the climatic conditions, hence abetting its decision of growing the suitable crop, under those factors. IoT grounded Smart Farming improves the entire husbandry system by covering the field in real- time. It keeps colorful factors like moisture, temperature, soiletc. under check and gives a crystalclear real- time observation. Machine literacy in husbandry is used to ameliorate the productivity and quality of the crops in the husbandry sector. Use of applicable algorithms on the tasted data can help in recommendation of suitable crop.

Introduction

I. INTRODUCTION

Agriculture plays a vital  part in the Indian frugality. Over  70 of the  pastoral  homes depend on  husbandry. Agriculture  is an important sector of Indian frugality as it contributes about  17 to the total GDP and provides employment to over 60 of  the population. Indian  husbandry has registered  emotional  growth over last many decades.  still,  growers ’  self-murders in India is  fussing. The  unexpressed reasons in order of  significance behind  planter suicides  were – debt,  terrain, low  yield prices, poor irrigation,  increased cost of  civilization, use of chemical diseases and crop  failure. A  planter’s decision about which crop to grow is  generally clouded by his suspicion and other  inapplicable factors  like making instant  gains, lack of  mindfulness about  request  demand,  overvaluing a soil’s  eventuality to support a particular  crop and so on. The need of the hour is to design a system that  could  give prophetic   perceptivity to the Indian  growers, thereby  helping them make an informed decision about which crop to  grow. This calls for the need of smart  husbandry, which requires  use of IoT. operation of IoT in  husbandry could be a life  changer for humanity and the whole earth. Detector data analytics  drives  translucency into agrarian processes, as  growers get  precious  perceptivity on the performance of their fields, glasshouses, etc. tilling powered by Machine Learning with  its high-  perfection algorithms is a new conception arising  moment.  Aiming to increase the  volume and quality of products, this   slice- edge movement makes sustainable productivity growth  for everyone working in the  husbandry realm.  With this in mind, we propose system for Smart  operation of Crop civilization using IoT and Machine  literacy – a smart system that can  help  growers in crop   operation by considering  tasted parameters( temperature,   moisture) and other parameters( soil type,  position of  ranch,   downfall) that predicts the most suitable crop to grow in that terrain.

II. LITERATURE REVIEW

As per Climate Change Crop Yield hypotheticals report in  [9], Crop Yields are projected to decline, with the larger declines  to be anticipated in several developing  husbandry which includes  Southeast Asia(- 5 percent) and India(- 5 percent). [10] The  variation in on-  ranch losses across regions may be  incompletely  explained by the range of reasons which include  structure  and marketing challenges,  infelicitous crop timing,  unanticipated  harsh climatic conditions and  unfit to  prognosticate suitable crops for   husbandry in  similar climates. The following comparison is show  below

Rushika Ghadge, Juilee Kulkarni, Pooja More, Sachee Nene,  Priya R L in [1] uses unsupervised and supervised  literacy  algorithms like Kohonen Self Organizing Map and Back  Propagation Network. Dataset is trained by learning networks to  classify it into organic, inorganic and real estate for  prognosticating  the type of soil. It compares the  delicacy  attained by different  network  literacy  ways and the most accurate result is  delivered to the end  stoner. System will check soil quality and   prognosticate the crop yield consequently along with it  give toxin  recommendation if  demanded depending upon the quality of soil.

Reference Paper [2] determines real time  slice of soil   parcels using MODIFIED SUPPORT VECTOR  REGRESSION, a popular machine learning algorithm and four  modules. The Modules include Sensor connived to IoT device,  Agri cloud, assaying the real time detector data and Agri  stoner  interface( AUI). The first module is  movable  IoT device  NodeMCU) with soil  humidity detector and pH detector,  environmental detectors. Agri cloud module consists of  storehouse.  assaying the real time data module is processing of types of  crops and small  shops suggested using modified support vector  machine algorithm. Agri-  stoner interface is a  introductory web interface.  therefore, with the help of soil  parcels  planter will be  suitable to get  types of crops and small  shops is grown in cropland with help  of Modified support vector machine algorithm.  18  International Journal of Engineering Research & Technology( IJERT)  ISSN 2278- 0181  Published by,www.ijert.org  NTASU- 2020 Conference Proceedings  Volume 9, Issue 3  Special Issue- 2021

[3] predicts temperature,  humidity and pH value for crop   vaticination using the ARIMA model. The model takes the values  from database as input and  also predicts what will be the value  of that particular parameter after 1 month. The  prognosticated values  are  also  transferred to K means algorithm for bracket grounded on  pH value  therefore creating k clusters of crops having  analogous pH  value. KNN algorithm is used to  prognosticate top N suitable crops  which are displayed to  stoner.

In [4], Machine Learning Algorithm( KNN) calculates the  parameter to suggest the crop which is stylish to grow in the  particular field grounded on the values  entered at real time. A  standardized dataset containing the  minimal conditions for a  particular crop is maintained and is used for the  vaticination of the  crop. The detectors are added to the field for which the readings  are  demanded to be calculated. The DHT11, MQ2, Soil humidity  Sensor, Light Intensity Sensor sends the readings in real time to  the  pall garçon. 

[5] evaluates the crop quality factor grounded onpre-established  rainfall conditions and nature of soil using the trained set of data  and  enforcing Supervised and underpinning models of  machine  literacy. If any  inimical conditions are perceived  ahead of time, the  volition and precautious measures are   espoused so as to  insure the good of the planted crops and  agrarian land. Specific measures are also taken to  prognosticate the  right period of sowing, reaping and  harvesting for the overall   improvement of the  product which can also be  previsioned as a  part of the  ultramodern agrarian revolution.

III. PROPOSED SYSTEM

The system aims to help  growers for smart decision  while  prognosticating the crops.  To increase the  delicacy along with  live data,  major data for temperature and  moisture from  government website is also collected and stored. Also  major   downfall data is collected and stored.  To be definite and accurate  in crop  vaticination, the  design analyzes the temperature and   moisture of the field –  live data collected using DHT- 22 detector  and  major data collected from government website and/ or  google rainfall API, type of the soil –  used by the  planter and  the  major  downfall data. It can be achieved using unsupervised  or supervised machine  literacy algorithm. By learning  networks, dataset is trained. The  delicacy  attained by different  machine   literacy  ways  are compared to get the  most  accurate result  which in turn  will be delivered to the  end  stoner.  Along  with the most suitable crop, the system also recommends  the toxin for that crop. Responsive, Multilingual website is  used by  planter to communicate with the system.

A. Hardware Components

Digital Temperature & moisture Detector DHT22  detector is preferred  to cover live temperature and  moisture.  This detector is proved to be more precise  and accurate. It uses a  capacitive  moisture detector and a thermistor to measure the   girding air, and spits out a digital signal on the data leg  to  Arduino Uno  harborage leg.  The range of DHT22 is 0 to- 100 RH  for  moisture and -  40 to 80 degree  Celsius  for temperature. 

Conclusion

In this paper, we\'ve proposed an innovative and comprehensive approach to smart husbandry that harnesses the important capabilities of two slice- edge technologies the Internet of effects( IoT) and Machine literacy( ML). By seamlessly integrating these two arising fields, our proposed system aims to revise the agrarian sector, empowering growers with unknown perceptivity and decision- support tools. The synergistic combination of live data aqueducts from IoT detectors and literal data depositories enables our system to achieve unequaled situations of delicacy in its prognostications and recommendations. By continuously covering real- time environmental conditions, soil humidity situations, rainfall patterns, and other pivotal factors, the IoT element provides a constant sluice of over- to- date information. contemporaneously, the integration of literal data allows our ML algorithms to identify patterns, trends, and correlations that may not be incontinently apparent, farther enhancing the system\'s prophetic capabilities. Likewise, our approach leverages the power of multiple ML algorithms, each with its unique strengths and specializations. By employing an ensemble of algorithms and using their reciprocal capabilities, our system can give further robust and accurate analyses, counting for the essential complications and variabilities present in agrarian disciplines. The ultimate thing of our proposed result is to palliate the challenges faced by growers and contribute to the sustainable growth and productivity of the agrarian sector. By furnishing data- driven perceptivity, optimized resource allocation strategies, and timely cautions for implicit pitfalls or openings, our system empowers growers to make informed opinions, maximize yields, and minimize destruction. also, our approach has the implicit to enhance the quality of agrarian yield, icing that crops are cultivated under optimal conditions and clinging to strict quality norms. This not only benefits growers economically but also contributes to food security and promotes environmentally sustainable practices. In summary, our innovative approach to smart husbandry, driven by the community of IoT and ML technologies, represents a significant stride towards a more effective, productive, and sustainable agrarian future. By employing the power of data, analytics, and slice- edge technologies, we aim to revise the way husbandry is rehearsed, fostering a paradigm shift that benefits growers, consumers, and the terrain likewise.

References

[1] Rushika Ghadge, Juilee Kulkarni, Pooja More, Sachee Nene, Priya R L, “Prediction of Crop Yield using Machine Learning,” International Research Journal of Engineering and Technology (IRJET) Feb 2018, pp. 2237-2239 [2] Radhika, Narendiran, “Kind of Crops and Small Plants Prediction using IoT with Machine Learning,” International Journal of Computer & Mathematical Sciences April 2018, pp. 93-97 [3] Shridhar Mhaiskar, Chinmay Patil, Piyush Wadhai, Aniket Patil, Vaishali Deshmukh, “A Survey on Predicting Suitable Crops for Cultivation Using IoT,” International Journal of Innovative Research in Computer and Communication Engineering January 2017, pp. 318-323 [4] T Raghav Kumar, Bhagavatula Aiswarya, Aashish Suresh, Drishti Jain, Natesh Balaji, Varshini Sankaran, “Smart Management of Crop Cultivation using IOT and Machine Learning,” International Research Journal of Engineering and Technology (IRJET) Nov 2018, pp. 845-850 [5] Akash Raj N, Balaji Srinivasan, Deepit Abhishek D, Sarath Jeyavanth J, Vinith Kannan A, “IoT based Agro Automation System using Machine Learning Algorithms”, International Journal of Innovative Research in Science, Engineering and Technology November 2016, pp.19938-19342 [6] Food and Agricultural Organization. (2018). The state of Agricultural Commodity Markets. 19-22. [7] Food and Agricultural Organization. (2019). The state of Food and Agriculture. 23-27.

Copyright

Copyright © 2024 Prof. A. A. Chaudhari, Rishi Vyas, Shivam Supekar, Kartik Dhole, Darshana Gujare, Rushikesh Chandekar. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.