Cover Image

Assessment of coconut tree crown detection using remote sensing approach / Nurul Annissa Abdullah

In this thesis, the identification of coconut tree crowns with the help of remote sensing is examined. The problem statement focuses on the difficulties in properly defining the class of coconut tree crowns for the purpose of difference in the image resolution and methods of classification. Thus, th...

Full description

Saved in:
Bibliographic Details
Main Author: Abdullah, Nurul Annissa
Format: Student Project
Language:English
Published: 2024
Subjects:
Remote Sensing
Online Access:https://ir.uitm.edu.my/id/eprint/109429/1/109429.pdf
https://ir.uitm.edu.my/id/eprint/109429/
Tags: Add Tag
No Tags, Be the first to tag this record!
id my.uitm.ir.109429
record_format eprints
spelling my.uitm.ir.1094292025-03-23T18:18:02Z https://ir.uitm.edu.my/id/eprint/109429/ Assessment of coconut tree crown detection using remote sensing approach / Nurul Annissa Abdullah Abdullah, Nurul Annissa Remote Sensing In this thesis, the identification of coconut tree crowns with the help of remote sensing is examined. The problem statement focuses on the difficulties in properly defining the class of coconut tree crowns for the purpose of difference in the image resolution and methods of classification. Thus, the objective of the study is to analyse the performance of a range of ultra metric measures of supervised classification methods such as SVM, RF, and MLC used in the analysis of the Sentinel 2 and SPOT 6 satellite imagery data. The specific objectives are as follows: to compare the effectiveness of these two methods in establishing the locations of coconut trees and other kinds of land covers. Chapter 4 further shows that for the MLC method the overall accuracy was 100 percent with Kappa coefficient of 1. 000 for the Sentinel 2A imagery this showing perfect classification. On the other hand, the latter or RF method aligned with the results that present a very low accuracy of 31. 5% and comically misplaced the classes. In the case of SPOT 6 imagery, both the MLC and RF classification exercise yielded impressive results with the accuracy levels of 92. 3% and the Kappa coefficient of 0. 859, thus this is an indication of the role played by image resolution and classification techniques in establishing high accuracy levels. The conclusion at chapter 5 is declaring that both SVM and MLC methods are more accurate than RF in the classification of tree crown on different satellite image data. Since correct classification of the land cover is significant, the study focuses on the relevance of having correct classification techniques and high-resolution images for the classification of the land cover, which will serve as a guide for further study and utilization in the agricultural sector. 2024-07 Student Project NonPeerReviewed text en https://ir.uitm.edu.my/id/eprint/109429/1/109429.pdf Assessment of coconut tree crown detection using remote sensing approach / Nurul Annissa Abdullah. (2024) [Student Project] (Submitted)
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
spellingShingle Remote Sensing
Abdullah, Nurul Annissa
Assessment of coconut tree crown detection using remote sensing approach / Nurul Annissa Abdullah
description In this thesis, the identification of coconut tree crowns with the help of remote sensing is examined. The problem statement focuses on the difficulties in properly defining the class of coconut tree crowns for the purpose of difference in the image resolution and methods of classification. Thus, the objective of the study is to analyse the performance of a range of ultra metric measures of supervised classification methods such as SVM, RF, and MLC used in the analysis of the Sentinel 2 and SPOT 6 satellite imagery data. The specific objectives are as follows: to compare the effectiveness of these two methods in establishing the locations of coconut trees and other kinds of land covers. Chapter 4 further shows that for the MLC method the overall accuracy was 100 percent with Kappa coefficient of 1. 000 for the Sentinel 2A imagery this showing perfect classification. On the other hand, the latter or RF method aligned with the results that present a very low accuracy of 31. 5% and comically misplaced the classes. In the case of SPOT 6 imagery, both the MLC and RF classification exercise yielded impressive results with the accuracy levels of 92. 3% and the Kappa coefficient of 0. 859, thus this is an indication of the role played by image resolution and classification techniques in establishing high accuracy levels. The conclusion at chapter 5 is declaring that both SVM and MLC methods are more accurate than RF in the classification of tree crown on different satellite image data. Since correct classification of the land cover is significant, the study focuses on the relevance of having correct classification techniques and high-resolution images for the classification of the land cover, which will serve as a guide for further study and utilization in the agricultural sector.
format Student Project
author Abdullah, Nurul Annissa
author_facet Abdullah, Nurul Annissa
author_sort Abdullah, Nurul Annissa
title Assessment of coconut tree crown detection using remote sensing approach / Nurul Annissa Abdullah
title_short Assessment of coconut tree crown detection using remote sensing approach / Nurul Annissa Abdullah
title_full Assessment of coconut tree crown detection using remote sensing approach / Nurul Annissa Abdullah
title_fullStr Assessment of coconut tree crown detection using remote sensing approach / Nurul Annissa Abdullah
title_full_unstemmed Assessment of coconut tree crown detection using remote sensing approach / Nurul Annissa Abdullah
title_sort assessment of coconut tree crown detection using remote sensing approach / nurul annissa abdullah
publishDate 2024
url https://ir.uitm.edu.my/id/eprint/109429/1/109429.pdf
https://ir.uitm.edu.my/id/eprint/109429/
_version_ 1827446782334861312
score 13.250246

Similar Items