Volume 10, Issue 2 (Autumn & Winter 2024)
KJES 2024, 10(2): 443-464 |
Back to browse issues page
Download citation:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
Babaeepour A, Milan A. Landuse change mapping based on drone and ultracam aerial images. KJES 2024; 10 (2) :443-464
URL: http://gnf.khu.ac.ir/article-1-2924-en.html
URL: http://gnf.khu.ac.ir/article-1-2924-en.html
1- Shahid Beheshti University , a.babaeepour@mail.sbu.ac.ir
2- Shahid Beheshti University
2- Shahid Beheshti University
Abstract: (185 Views)
The use of drone imagery has become widespread in studies related to various fields, including urban planning. The application of drone photogrammetry as an advanced tool for analyzing urban and construction changes is considered a novel method in spatial-based research. In this study, the potential of drone imagery, specifically DJI Phantom 4 Pro RTK with a relative accuracy of 10 cm, alongside ultra-cam aerial images with a relative accuracy of 20 cm, was explored for identifying and analyzing land-use changes in densely urbanized areas of the Tabriz region. To achieve this, two data sets were utilized, including a corrected orthomosaic image from the ultra-cam camera captured on June 9, 2013, and raw drone images from the same area on July 4, 2022. After geometric correction and the creation of orthophotos from the raw drone images, with horizontal accuracy of 8 cm and vertical accuracy of 14 cm, two classification algorithms, maximum likelihood and minimum distance, were applied to the orthomosaic image and the drone orthophoto. The next step involved using the thematic change detection method to extract land-use changes in the identified classes based on the two classification algorithms. Visual evaluation of the results revealed that the building class experienced the least change compared to other classes. Quantitative findings showed that the Kappa coefficient and overall accuracy for the maximum likelihood and minimum distance classification methods were 0.8924 and 94.17%, and 0.5273 and 93.08%, respectively. Additionally, quantitative analysis indicated that the greatest land-use change involved the conversion of buildings to roads, while the least change occurred in the transformation from roads to barren land.
Type of Study: Original Research |
Subject:
Remote sensing and GIS
Received: 2024/10/8 | Accepted: 2024/12/21 | Published: 2024/12/28
Received: 2024/10/8 | Accepted: 2024/12/21 | Published: 2024/12/28
Send email to the article author
| Rights and permissions | |
 |
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. |
