Bhoi Ashutosh, Nayak Rajendra Prasad, Bhoi Sourav Kumar, Sethi Srinivas, Panda Sanjaya Kumar, Sahoo Kshira Sagar, Nayyar Anand
Department of Computer Science and Engineering, Government College of Engineering (Govt.), Kalahandi, India.
Department of Computer Science and Engineering, Parala Maharaja Engineering College (Govt.), Berhampur, India.
PeerJ Comput Sci. 2021 Jun 21;7:e578. doi: 10.7717/peerj-cs.578. eCollection 2021.
In the traditional irrigation process, a huge amount of water consumption is required which leads to water wastage. To reduce the wasting of water for this tedious task, an intelligent irrigation system is urgently needed. The era of machine learning (ML) and the Internet of Things (IoT) brings it is a great advantage of building an intelligent system that performs this task automatically with minimal human effort. In this study, an IoT enabled ML-trained recommendation system is proposed for efficient water usage with the nominal intervention of farmers. IoT devices are deployed in the crop field to precisely collect the ground and environmental details. The gathered data are forwarded and stored in a cloud-based server, which applies ML approaches to analyze data and suggest irrigation to the farmer. To make the system robust and adaptive, an inbuilt feedback mechanism is added to this recommendation system. The experimentation, reveals that the proposed system performs quite well on our own collected dataset and National Institute of Technology (NIT) Raipur crop dataset.
在传统灌溉过程中，需要大量的水消耗，这导致了水资源的浪费。为了减少这项繁琐任务中的水资源浪费，迫切需要一个智能灌溉系统。机器学习（ML）和物联网（IoT）时代为构建一个智能系统带来了巨大优势，该系统能够以最少的人力自动执行此任务。在本研究中，提出了一种基于物联网的经过ML训练的推荐系统，用于在农民的名义干预下高效用水。物联网设备部署在农田中，以精确收集地面和环境细节。收集到的数据被转发并存储在基于云的服务器中，该服务器应用ML方法分析数据并向农民建议灌溉。为了使系统健壮且自适应，在这个推荐系统中添加了一个内置反馈机制。实验表明，所提出的系统在我们自己收集的数据集和印度国家技术学院（NIT）赖布尔作物数据集上表现良好。
