International Journal of Science and Research (IJSR)

International Journal of Science and Research (IJSR)
Call for Papers | Fully Refereed | Open Access | Double Blind Peer Reviewed

ISSN: 2319-7064

Downloads: 108 | Views: 277 | Weekly Hits: ⮙2 | Monthly Hits: ⮙2

Research Paper | Earth Science and Engineering | India | Volume 6 Issue 5, May 2017 | Rating: 6.8 / 10


Fire Detection Using Support Vector Machines (SVM)

Vigneshwaran SR | S. S. Shanthakumari | Vinodhini Ranganathan


Abstract: Earth is an intrinsically flammable planet owing to its cover of carbon-rich vegetation, seasonally dry climates, wide spread lightning and volcano ignitions. Therefore, something is always burning on this earth. Wildfires can be natural or man-made (accidental or deliberate). These fires release greenhouse gases and huge volumes of smoke thereby polluting the air and the environment which in turn cause severe degradation to the ecosystems. Based on observations from the Moderate Resolution Imaging Spectroradiometer (MODIS) on NASAs Terra and Aqua satellites, several algorithms have been designed and many approaches have been proposed to detect wildfire. While the question of accuracy over the existing methods of fire detection remains, this paper proposes a more robust and accurate approach using SVM (Support Vector Machines) for detecting wildfire. And also will compare SVM with logistic regression which is the most common used algorithm for solving industry problems.


Keywords: SVM, wildfire, MODIS


Edition: Volume 6 Issue 5, May 2017,


Pages: 1607 - 1618





Rate this Article


Select Rating (Lowest: 1, Highest: 10)

5

Your Comments

Characters: 0

Your Full Name:


Your Valid Email Address:


Verification Code will appear in 2 Seconds ... Wait

Top