期刊名称：Journal of Theoretical and Applied Information Technology
印刷版ISSN：1992-8645
电子版ISSN：1817-3195
出版年度：2019
卷号：97
期号：3
页码：879-896
出版社：Journal of Theoretical and Applied
摘要：The staple food for the Indonesian people is rice because in rice it contains a large number of calories for the intake of more than 200 million people. Hyperspectral is the sensors that can be used in a variety of applications, one of which is for rice monitoring. Hyperspectral is a sensor that is very well used to support precision agriculture because the information obtained is more detailed. One method of monitoring rice is to use a classification method. Many classification methods were carried out in previous hyperspectral studies such as unsupervised, supervised, statistical-based and so forth. Some methods have their own advantages and disadvantages. However, hyperspectral imagery has a large number of bands, requires sophisticated analytical methods to analyze it and requires a long process to extract priority information so as not to burden computing. In this paper discusses the state-of-the-art framework and step by step regarding the classification methods commonly used to rice monitoring. From the results of the review, it was found that the RBFN classification technique has the best accuracy compared to other classification techniques.
