Author:
Phasinam Khongdet,Kassanuk Thanwamas
Abstract
A human's ability to survive would be impossible without agriculture. Agriculture provides a means of subsistence for a huge proportion of the global population. It also provides a large number of work opportunities for the locals. Poor yields are a consequence of farmers' desire to rely on old-fashioned farming methods. The long-term development and success of the economy will continue to depend on agricultural and related industries. Crop monitoring and tracking, disease identification and management, and other such issues are among agriculture's most difficult difficulties. In many cases, smart farming is a viable answer. The Internet of Things (IoT) and machine learning methods has made it feasible to implement smart agriculture. This article presents a framework for real-time picture categorization in agriculture. Everything from IoT cameras to mobile applications to machine learning methods is featured. Arduino Uno, sensors, and Wi-Fi devices make up the hardware. Computers may "learn" from previous examples and detect patterns from noisy or complicated datasets using machine learning, a technique that uses a variety of statistical, probabilistic, and optimization methodologies to allow computers to "learn." Because of this, machine learning algorithms are increasingly being utilized to classify photos. Real-time photos linked to smart agriculture were classified using a combination of SVM, K-nearest neighbors, and probabilistic neural network classifiers. Real-time picture classification will assist in the diagnosis of leaf disease, the monitoring of farm staff, the classification of crops, and the tracking of farm product progress.
Publisher
The Electrochemical Society
Cited by
4 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献