Cover Image

Building extraction of worldview3 imagery via support vector machine using scikit-learn module / Najihah Ismail

Building is one of the major features that available on the land used on the earth especially in urban area. By using remote sensing method, it can reduce time to collect data for a large area. The data can be gain in high or low resolution. Low resolution image is cheaper and easy to access meanwhi...

Full description

Saved in:
Bibliographic Details
Main Author: Ismail, Najihah
Format: Thesis
Language:English
Published: 2021
Subjects:
Remote Sensing
Land use
Online Access:http://ir.uitm.edu.my/id/eprint/42992/1/42992.pdf
http://ir.uitm.edu.my/id/eprint/42992/
Tags: Add Tag
No Tags, Be the first to tag this record!
id my.uitm.ir.42992
record_format eprints
spelling my.uitm.ir.429922021-03-10T01:09:07Z http://ir.uitm.edu.my/id/eprint/42992/ Building extraction of worldview3 imagery via support vector machine using scikit-learn module / Najihah Ismail Ismail, Najihah Remote Sensing Land use Building is one of the major features that available on the land used on the earth especially in urban area. By using remote sensing method, it can reduce time to collect data for a large area. The data can be gain in high or low resolution. Low resolution image is cheaper and easy to access meanwhile high resolution can provide a better view and more accurate to differentiate the features available on image. Nowadays, researchers have investigated the use of different approach for building classification and extraction. However, there is need to monitor the effectiveness of the method used effectively. Consequently, this study is intending to apply Support Vector Machine (SVM) classification which using Scikit-learn module for building classification. Moreover, the capability of the programming based using python for building extraction can be assessed. Python is an open source of programming software that conducted programming-based technique using the Scikit-Learn module to do the extraction of building from Land used land cover (LULC) and the result was 86.233% for overall accuracy. A Commercial Remote Sensing Technology (ENVI) was used and measured to improve and verify the performance of the Python programming-based picture classification by applying the same SVM algorithm and the tests indicated 95.0732% for an overall accuracy. 2021-03-05 Thesis NonPeerReviewed text en http://ir.uitm.edu.my/id/eprint/42992/1/42992.pdf Ismail, Najihah (2021) Building extraction of worldview3 imagery via support vector machine using scikit-learn module / Najihah Ismail. Degree thesis, Universiti Teknologi Mara Perlis.
institution Universiti Teknologi Mara
building Tun Abdul Razak Library
collection Institutional Repository
continent Asia
country Malaysia
content_provider Universiti Teknologi Mara
content_source UiTM Institutional Repository
url_provider http://ir.uitm.edu.my/
language English
topic Remote Sensing
Land use
spellingShingle Remote Sensing
Land use
Ismail, Najihah
Building extraction of worldview3 imagery via support vector machine using scikit-learn module / Najihah Ismail
description Building is one of the major features that available on the land used on the earth especially in urban area. By using remote sensing method, it can reduce time to collect data for a large area. The data can be gain in high or low resolution. Low resolution image is cheaper and easy to access meanwhile high resolution can provide a better view and more accurate to differentiate the features available on image. Nowadays, researchers have investigated the use of different approach for building classification and extraction. However, there is need to monitor the effectiveness of the method used effectively. Consequently, this study is intending to apply Support Vector Machine (SVM) classification which using Scikit-learn module for building classification. Moreover, the capability of the programming based using python for building extraction can be assessed. Python is an open source of programming software that conducted programming-based technique using the Scikit-Learn module to do the extraction of building from Land used land cover (LULC) and the result was 86.233% for overall accuracy. A Commercial Remote Sensing Technology (ENVI) was used and measured to improve and verify the performance of the Python programming-based picture classification by applying the same SVM algorithm and the tests indicated 95.0732% for an overall accuracy.
format Thesis
author Ismail, Najihah
author_facet Ismail, Najihah
author_sort Ismail, Najihah
title Building extraction of worldview3 imagery via support vector machine using scikit-learn module / Najihah Ismail
title_short Building extraction of worldview3 imagery via support vector machine using scikit-learn module / Najihah Ismail
title_full Building extraction of worldview3 imagery via support vector machine using scikit-learn module / Najihah Ismail
title_fullStr Building extraction of worldview3 imagery via support vector machine using scikit-learn module / Najihah Ismail
title_full_unstemmed Building extraction of worldview3 imagery via support vector machine using scikit-learn module / Najihah Ismail
title_sort building extraction of worldview3 imagery via support vector machine using scikit-learn module / najihah ismail
publishDate 2021
url http://ir.uitm.edu.my/id/eprint/42992/1/42992.pdf
http://ir.uitm.edu.my/id/eprint/42992/
_version_ 1695534638657699840
score 13.211869

Similar Items