Pixel- and Object-Based Image Analysis for Sentinel-2 LULC Classification: A Case Study in Yalova, Türkiye
Journal of the Indian Society of Remote Sensing, 2026 (SCI-Expanded, Scopus)
- Yayın Türü: Makale / Tam Makale
- Basım Tarihi: 2026
- Doi Numarası: 10.1007/s12524-026-02532-9
- Dergi Adı: Journal of the Indian Society of Remote Sensing
- Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Geobase, INSPEC, Natural Science Collection (ProQuest), Earth, Atmospheric, & Aquatic Science Collection (ProQuest), Materials Science & Engineering Collection (ProQuest), Technology Collection (ProQuest)
- Anahtar Kelimeler: LULC, Pixel-based and object-based classification, Sentinel-2, Yalova
- İstanbul Üniversitesi Adresli: Evet
Özet
Land use and land cover (LULC) on our planet is continuously changing due to both natural events and human activities. Monitoring these changes is essential for urban development and natural resource management. Recent developments in satellite image classification have greatly supported such applications, which require accurate data for the area of interest. In this study, Sentinel-2 imagery of Yalova was used to compare pixel-based and object-based LULC mapping methods using open-source (QGIS) and commercial (eCognition) software. The region is rapidly developing in terms of urbanization, infrastructure, and industry and is surrounded by agricultural and forest areas. Additionally, it is located within the seismically active North Anatolian Fault Zone. The classification analysis showed that pixel-based methods achieved an overall accuracy of nearly 96%. Object-based approaches produced slightly lower but still comparable results. Vegetation and waterbody classes reached accuracies of about 98–99%. Built-up areas and roads had lower accuracies, ranging between 55% and 70%. These results indicate that both approaches perform similarly for homogeneous land cover classes. Object-based methods with an optimized feature set perform better in identifying complex land cover types. This study highlights the importance of selecting appropriate classification strategies for LULC mapping in dynamic and disaster-prone regions.