Open AccessA Research Framework for Supervised Image Classification For Tornado Chaos Phenomena
Author(s) -
W. Wanayumini,
Opim Salim Sitompul,
Muhammad Zarlis,
Saib Suwilo,
Akim Manaor Hara Pardede
Publication year - 2018
Publication title -
international journal of engineering and technology
Language(s) - English
Resource type - Journals
ISSN - 2227-524X
DOI - 10.14419/ijet.v7i4.15.25254
Subject(s) - tornado , contextual image classification , computer science , chaotic , similarity (geometry) , chaos (operating system) , remote sensing , pixel , data mining , image (mathematics) , artificial intelligence , pattern recognition (psychology) , geology , meteorology , geography , computer security
Unattended classification is a classification which is the process of forming classes conducted by computers. The classes formed in this classification are highly dependent on data acquisition. In the process, this classification classifies pixels based on similarity or spectral similarity. While the supervised classification is a classification carried out by the analyst's direction. The purpose of this study is to build a new model of image-based classification based on chaos phenomena through remote sensing to detect the beginning of the emergence of tornadoes. This research optimizes the search for the best value from a data collection of samples of chaos phenomena in tornadoes through a new model called Citra which is supervised by Chaos Discrete Cosine Transform Spectral Angel Mapper Classification (SiChDCosTSamC). The resulting model can then be used as remote sensing to detect the appearance of the initial tornado. Tests will be carried out using the Protected Image Welding on models based on chaotic / chaotic phenomena. Testing will be carried out on a collection of sample image data sourced from SIO, NOAA, US data. Navy, NGA, GEBCO U.S. PGA / NASA Google IBCAO Geological Geological Survey / Copernicus.