DSpace logo

Please use this identifier to cite or link to this item: http://142.54.178.187:9060/xmlui/handle/123456789/1442
Title: Deforestation Analysis of Northern Areas(Pakistan) using Image Processing and Maximum Likelihood Supervised Classification
Authors: Rehman, Abdul
Shahid, Husnain
Sultan, Azeem
Afzal, Muhammad Abrar
Tahir, Muhammad Owais
Keywords: Engineering and Technology
Vegetation mapping
Agriculture
Floods
Geographic information systems
Geophysical image processing
Terrain mapping
Vegetation
Image classification
Issue Date: 6-Mar-2019
Publisher: IEEE 2nd International Conference on Communication, Computing and Digital systems (C-CODE)
Abstract: An extensive uncertainty is a big concern of deforestation, degradation and forest decentralization. Recorded cases of deforestation in northern regions of Pakistan have drawn serious involvement in the last two decades. These areas include agriculture land and source of fresh water for more than 20 million residents. Downgrading in the forest is also big damage in the ecosystem which increases the flood risk in any community. The fast development in remote sensing (RS) satellites and RS techniques in the last four decades, provides a stable, successive and efficient way for analysis of land cover and land mapping. This work is carried out the analysis of deforestation from 2000-2016. The satellite images of Landsat 4, 5 and 8 are used and processed by Arc GIS software. Maximum Likelihood (ML) supervised classification based on Bayes theorem is used and this technique classified the vegetation, water, and mountains individually. Normalized Difference Vegetation Index (NDVI) is used to calculate the downfall in forests from the vegetation area. This analysis reflected a compelling downturn in forest cover in a period of study.
URI: http://142.54.178.187:9060/xmlui/handle/123456789/1442
ISBN: 978-1-5386-9609-5
Appears in Collections:Proceedings

Files in This Item:
File Description SizeFormat 
8680986.htm115 BHTMLView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.