Sürdürülebilir Kalkınma Amaçları
Yayınlar
- MakaleYalnız künye2026Investigating class granularity in historical aerial image segmentation: a comparative analysis of CNN and transformer-based models
Dihkan, Mustafa; Akbulut, Zeynep; Güler, Gülsena; Özdemir, Samed; Karsli, Fevzi
Historical aerial imagery acts as a crucial bridge to the pre-satellite era, becoming increasingly vital for quantifying long-term land cover changes and supporting retrospective environmental studies. This study presents a comprehensive comparison of CNN-based (DeepLabV3+, PSPNet, HRNet, ConvNeXt) and transformer-based (ViT, Swin Transformer, BEiT) architectures for semantic segmentation of historical aerial imagery. In particular, the potential of Transformer-based architectures for semantic segmentation of panchromatic historical aerial imagery in the context of increasing segmentation complexity remains underexplored. We assess how effectively each architecture handles increasing class granularity and captures finer-grained land-cover semantics. Experiments are conducted at three levels of detail: binary, three-class, and five-class segmentation. The dataset comprises panchromatic images from 1960, captured in Turkey. Precision, recall, F1-score, and intersection-over-union (IoU) metrics were employed to evaluate class-wise performance. In the binary ground and non-ground segmentation task, all architectures achieved mean F1-scores of 0.945–0.959 and mean IoU values of 0.905–0.921. In the three-class and five-class tasks, mean F1-scores ranged from 0.861–0.902 and 0.754–0.803, with mean IoU values of 0.777–0.831 and 0.623–0.682, respectively. In the three-class setting, ConvNeXt, Swin Transformer, and BEiT clustered towards the upper end of the mean metrics with only small differences among them. Increasing class granularity from three to five reduced mean performance across architectures, with decreases of up to 0.123 in mean F1-score and up to 0.176 in mean IoU. Across all classification levels, ConvNeXt consistently achieved higher performance, distinguishing itself specifically in the five-class task with the highest F1-scores and IoU values. ConvNeXt outperformed both the CNN-based baselines and Transformer-based models in the challenging building and road categories.
- MakaleYalnız künye2025Vision Foundation Models and Rule-Based Approaches for Roof Surface Segmentation and Photovoltaic Potential Analysis in Urban Areas
Özdemir, Samed; Yavuzdoğan, Ahmet
This study presents two methods for rapidly and effectively determining the photovoltaic (PV) potential of building roofs in urban areas using aerial photographs and point cloud data. In the first method, the Segment Anything Model (SAM) and Contrastive Language Image Pre-Training (CLIP) models are used to detect roof surfaces and obstacles from aerial photographs. In the second method, the Random Sample Consensus (RANSAC) and Density-Based Spatial Clustering of Applications with Noise (DBSCAN) algorithms are employed to identify roof surfaces from Light Detection and Ranging (LiDAR) point clouds. Through the first proposed method, the performance of current deep learning approaches in 2.5D PV potential analysis is investigated, while the second approach examines the performance of 3D PV potential analysis compared to the 2D approach. In PV potential analysis, the Photovoltaic Geographical Information System (PVGIS) Application Programming Interface (API) was utilized. The analysis is conducted based on roof parameters obtained through both proposed methods. In building detection, the first approach achieved an Intersection over Union (IoU) score of 94.29%, whereas the second approach attained an IoU score of 91.23%.
- BildiriAçık Erişim2025Comparative Analysis of Vision Foundation Models For Building Segmentation in Aerial Imagery
Akbulut, Zeynep; Özdemir, Samed; Karslı, Fevzi; Sefercik, U.G.
Visual Foundation Models (VFMs) demonstrate impressive generalization capabilities for image segmentation and classification tasks, leading to their increasing adoption in the remote sensing field. This study investigates the performance of VFMs in zero-shot building segmentation from aerial imagery using two model pipelines: Grounded-SAM and SAM+CLIP. Grounded-SAM integrates the Grounding DINO backbone with a Segment Anything Model (SAM) while SAM+CLIP first employs SAM for generating masks followed by Contrastive Language Image Pretraining (CLIP) for classification. The evaluation, performed on the WHU building dataset using Precision, Recall, F1 score, and intersection over union (IoU) metrics, revealed that Grounded-SAM achieved F1-score of 0.83 and IoU of 0.71. SAM+CLIP achieved F1-score of 0.65 and IoU of 0.49. While Grounded-SAM excelled at accurately delineating partially occluded and irregularly shaped buildings, SAM+CLIP was able to segment larger buildings but struggled with delineating smaller ones. Given the impressive performance of VFMs in zero-shot building segmentation, future efforts aimed at refining these models through fine-tuning or few-shot learning could significantly expand their application in remote sensing.
- MakaleAçık Erişim2024Nokta bulutu verisi ile su kütlesi tespitinde geometrik özniteliklerin etkisi
Özdemir, Samed; Karsli, Fevzi
Yüksek çözünürlüklü uzaktan algılama görüntülerinden su kütlelerinin hızlı ve doğru bir şekilde tespit edilmesi ve çıkarılması, su kaynakları yönetimi, su kalitesi izleme, doğal afet acil müdahaleleri gibi kritik uygulama alanlarında stratejik bir öneme sahiptir. Bununla birlikte, geleneksel su kütle çıkarma yöntemleri, özellikle görüntü dokusu ve karakteristik özelliklerin seçilmesi konusunda çeşitli zorluklar sunmaktadır. Bu çalışmada, nokta bulutu verilerinden çıkarılan geometrik öznitelikler ve hava fotoğraflarından alınan spektral bilgileri bir araya getirerek, su kütlelerinin sınırlarının daha etkin bir şekilde tanımlanmasını ve çıkarılmasını sağlayan bir metodoloji önerilmektedir. Üç boyutlu (3B) yapı tensöründen yararlanılarak nokta bulutlarından üretilen geometrik öznitelikler algılayıcı sistemin ürettiği spektral bilgiler ile birleştirilerek, yüksek boyutlu verilere uygunluğu, hızı ve aşırı uyuma direnci ile bilinen Rastgele Orman (RO) sınıflandırıcısı su kütlelerinin tespiti için eğitimde kullanılmıştır. Matlab ortamında geliştirilen metodolojinin etkinliği, Türkiye’de topografik ve bitkisel özellikleri farklı dört farklı lokasyon üzerinde test edilmiştir. Sınıflandırma işlemi ile tespit edilen su kütlesi sınırlarının doğruluk analizi F-Skoru üzerinden değerlendirildiğinde, Çalışma Alanı-1 için: %85.7, Çalışma Alanı-1 Akarsu için %76.6, Çalışma Alanı-2 için %93.7, Çalışma Alanı-3 için %94.9, ve Çalışma Alanı-4 için %73.6, olarak elde edilmiştir. Çalışma, sunulan metodolojinin farklı mekânsal ölçekler ve sensör türleri için uygulanabilir olduğunu ve çevresel ve hidrolojik araştırmalarda geniş kapsamlı kullanımlar için potansiyel taşıdığını ortaya koymaktadır.
- MakaleAçık Erişim20233D data integration for geo-located cave mapping based on unmanned aerial vehicle and terrestrial laser scanner data
Çömert, Resul; Özdemir, Samed; Bilgilioglu, Burhan Baha; Alemdag, Selçuk; Zeybek, Halil Ibrahim
The Akçakale cave is a significant natural and cultural heritage site in the Black Sea region of eastern Turkey. The complex geometry and difficult-to-access areas of the cave have made the use of traditional mapping methods challenging. To overcome these limitations, this study utilized TLS and UAV technology to produce highly accurate 2D and 3D data for cave management and risk assessment purposes. The TLS system was used to create a detailed 3D point cloud of the cave interior, while the UAV system generated a 3D model of the surface topography outside the cave. The two sets of data were combined in the GIS environment using a geodetic network established in the study area, providing a common geodetic reference system for both TLS and UAV data. The study found that the cave area is 13,750 m2, which is smaller than the area of 18,000 m2 that was previously estimated using conventional measurement methods. The volume and ceiling heights of the cave were calculated using the elevation models generated from TLS point cloud data. The 3D point cloud data were also used to map dripstone locations on the floor and ceiling of the cave, and the boundaries of rock blocks on the ground were precisely determined. The study identified potential risks associated with the cave, particularly the risk of rockfall in the source rock areas around the cave entrance and the southern part of the cave. The nearest building to the cave is approximately 35 meters away, and all the buildings in the area are less than 300 meters from the cave. In the event of the cave collapse, the buildings in the southern part of the cave are at risk of rockfall. This study demonstrates the effectiveness of combining data from TLS and UAV systems to generate broad and sensitive cave mapping and risk assessment data, which are critical for cave management and safety. The collected data can be used for cave stability investigations and rockfall risk assessments. This study provides a foundation for future explorations of the Akcakale cave and highlights the potential for modern surveying techniques to enhance our understanding of complex geological structures such as caves.
- MakaleYalnız künye2023SPAN: An open-source plugin for photovoltaic potential estimation of individual roof segments using point cloud data
Özdemir, Samed; Yavuzdoğan, Ahmet; Bilgilioğlu, Burhan Baha; Akbulut, Zeynep
Decentralized solar PhotoVoltaic (PV) is one of the most promising energy sources for cities and individuals pursuing energy self-sufficiency. Especially, the already available rooftop surfaces are a major contributor to push for rooftop mounted PV systems. However, accurate PV potential estimation of individual buildings is still a challenging task since many parameters must be considered such as meteorological factors, panel technology, geographical location, available roof surface area, surface azimuth and tilt angle. In this study, we created an efficient approach that can be used for roof surface's PV potential estimation based on point cloud data and capable of processing various scales from single building to city scale. In the proposed approach, each roof surface's features were utilized for PV potential estimation by employing the PVGIS database. PV potential estimation was carried out on daily, monthly, and annual periods to provide a better estimation. Also, we developed a flexible and easy to use open-source plugin based on the QGIS software for rooftop mounted PV potential estimation capable of estimating every roof surface's PV potential. The method was tested on 80 buildings selected from ROOFN3D dataset. The proposed approach achieved an overall accuracy of 84% and an F1 score of 0.92.
- MakaleYalnız künye2018Automatic Building Extraction from Image and LiDAR Data with Active Contour Segmentation
Akbulut, Zeynep; Özdemir, Samed; Acar, Hayrettin; Karsli, Fevzi
Automatic building extraction is an important topic for many applications such as urban planning, disaster management, 3D building modeling and updating GIS databases. Its approaches mainly depend on two data sources: light detection and ranging (LiDAR) point cloud and aerial imagery both of which have advantages and disadvantages of their own. In this study, in order to benefit from the advantages of each data sources, LiDAR and image data combined together. And then, the building boundaries were extracted with the automated active contour algorithm implemented in MATLAB. Active contour algorithm uses initial contour positions to segment an object in the image. Initial contour positions were detected without user interaction by a series of image enhancements, band ratio and morphological operations. Four test areas with varying building and background levels of detail were selected from ISPRS’s benchmark Vaihingen and Istanbul datasets. Vegetation and shadows were removed from all the datasets by band ratio to improve segmentation quality. Subsequently, LiDAR point cloud data was converted to raster format and added to the aerial imagery as an extra band. Resulting merged image and initial contour positions were given to the active contour algorithm to extract building boundaries. In order to compare the contribution of LiDAR to the proposed method, the boundaries of the buildings were extracted from the input image before and after adding LiDAR data to the image as a layer. Finally extracted building boundaries were smoothed by the Awrangjeb (Int J Remote Sen 37(3): 551–579. https://doi.org/10.1080/01431161.2015.1131868, 2016) boundary regularization algorithm. Correctness (Corr), completeness (Comp) and accuracy (Q) metrics were used to assess accuracy of segmented building boundaries by comparing extracted building boundaries with manually digitized building boundaries. Proposed approach shows the promising results with over 93% correctness, 92% completeness and 89% quality.