Land Use/Land Cover Change in Manaus, Brazil

Presentation Type

Oral and/or Visual Presentation

Presenter Major(s)

Geography and Planning

Mentor Information

Wanxiao Sun

Department

Geography and Planning

Location

Kirkhof Center 2270

Start Date

11-4-2012 4:30 PM

Keywords

Environment, Sustainability, World Perspective

Abstract

The city of Manaus in Amazonia was the center of the 19th century rubber boom. Becoming the richest city of the Amazon, Manaus enjoys a rich history and expanding population. This study aims to show how the physical landscape has changed between 1986 and 2001 using Object-Oriented Image Segmentation and Classification with Landsat5-7 imagery. The process of classification works to determine land features (e.g. urban, forest, field, water, etc...) and can be used to find trends and statistics of land cover/land use change. The image data will be georectified and have atmospheric effects removed. The Object-Oriented Image Segmentation and Classification works by analysis of image objects and spatial relationships instead of on single pixels (i.e. traditional image classification). By creating rule sets, land cover can be extracted and classified more accurately.

This document is currently not available here.

Share

COinS
 
Apr 11th, 4:30 PM

Land Use/Land Cover Change in Manaus, Brazil

Kirkhof Center 2270

The city of Manaus in Amazonia was the center of the 19th century rubber boom. Becoming the richest city of the Amazon, Manaus enjoys a rich history and expanding population. This study aims to show how the physical landscape has changed between 1986 and 2001 using Object-Oriented Image Segmentation and Classification with Landsat5-7 imagery. The process of classification works to determine land features (e.g. urban, forest, field, water, etc...) and can be used to find trends and statistics of land cover/land use change. The image data will be georectified and have atmospheric effects removed. The Object-Oriented Image Segmentation and Classification works by analysis of image objects and spatial relationships instead of on single pixels (i.e. traditional image classification). By creating rule sets, land cover can be extracted and classified more accurately.