Jul 2026· 2026 4th International Conference on Sustainable Computing and Smart Systems (ICSCSS)· pp. 1984-1990· 0 citations· 14 references
Abstract
In the present generation of increasing geospatial data, accurate and automated extraction of building footprints from high-resolution aerial and satellite imagery has become crucial for various applications such as urban planning, infrastructure development, disaster management, and GIS database maintenance, as manual tracing is time-consuming and unstable for large-scale mapping. This study compares conventional image processing techniques such as thresholding, edge detection, morphological operations through a machine learning approach using Random Forest (RF), and deep learning-based semantic segmentation models, namely U-Net and DeepLabV3+, along with the Segment Anything Model (SAM) using a pre-trained prompt-based setup. All methods are tested on the same set of data, and a standardized data preprocessing is performed for fair comparison. The overall results indicate that the application of DeepLabV3+ is best, with an IoU of 82% and an F1 score of 90%. U-Net achieves second high IoU and F1 scores of 74% and 84% respectively, while Random Forest shows a high IoU of 60% and an F1-score of 72%. SAM has the lowest scores with an IoU of 50% and an F1 score of 51%.
Abstract. Extracting road networks from high-resolution imagery remains a significant challenge in geomatics, particularly in fragmented rural landscapes. The big difficulty is the spectral similarities between unpaved tracks and agricultural backgrounds that can lead to classification errors. This study proposes an automated geospatial pipeline based on the YOLOv11 architecture. Specifically, the approach is made on the optimization of the multi-class road detection in the rural areas of Kosina and Markowa, two villages in Poland. To reduce the computational effort, due to large-scale 9000×9000 px orthophotos and to improve the detection of small-scale features, Slicing Aided Hyper Inference (SAHI) strategy was integrated. High-resolution imagery has been decomposed into optimized tiles, ensuring feature continuity across boundaries and preventing GPU memory overhead. The instance segmentation model was trained on a custom-annotated dataset, with seven labels (categories) such as internal paved roads, rural tracks, and railway infrastructures. Therefore, a high level of robustness has been achieved reaching a mean Average Precision value (mAP@0.5) of 0.90. A confusion matrix reveals quantitatively that the pipeline effectively distinguishes between complex classes and low omission rates. As a result, the generated outputs are converted into interoperable GeoJSON format ensuring their integration into GIS environments. In conclusion, the experimental result demonstrates that the framework is valuable for emergency response logistics and urban planning. It offers a scalable and near real-time solution for updating national topographic databases.
M. Sanità, L. Nepi, E. Malinverni et al.· The International Archives o...· 0 citations
SRB-Net is presented, a U-Net-based framework that combines three complementary components: strip pooling for long-range horizontal and vertical context; residual multi-scale atrous spatial pyramid pooling with squeeze-and-excitation blocks for multi-scale and channel-aware feature learning; and a bottleneck attention module (BAM) for refining skip-connection features.
Hamdoun Youssef, Xingyuan Li, Yongtao Yu et al.· Journal of Computing and Ele...· 0 citations
This work presents a new method for using the VFM Segment Anything Model 2 (SAM 2) for multi-class semantic segmentation of Sentinel-2 images that does not require training data and achieves an overall accuracy of up to 93% at pixel-level using polygon mask prompts.
Paula L. Lippmann, M. Dorozynski, F. Rottensteiner et al.· The International Archives o...· 0 citations
Building object extraction plays a crucial role in the monitoring and strategic planning of urban infrastructures. With the rapid advancement of deep learning (DL), segmentation tasks are becoming more effective, even though models trained on datasets from different regions may not perform well in Indian areas due to their complex, unstructured cityscapes. Therefore, this paper introduces a novel building object segmentation dataset, namely, NITK Computer Vision and Deep Learning (NCVDL) Cartosat-2S building dataset, which covers three diverse regions of India. Moreover, this paper presents a lightweight DL model, namely, the KAN Building Detection Network (KANBDNet), for extracting buildings from satellite images, aiming to reduce the number of parameters without compromising performance using Kolmogorov–Arnold Networks, and the training is guided by a boundary-focused mIoU metric. It is mainly structured using three modules: a split-wise feature extraction module to efficiently extract features by splitting them into two and then process separately, a dense multiscale feature extraction block to extract and densely connect features at various scales, and a skip layer which effectively combines the encoder stage and high level features to enhance the features passing through the decoder. The performance of the proposed KANBDNet and state-of-the-art (SOTA) DL-based building segmentation models is evaluated using the new NCVDL Cartosat-2S dataset and the publicly available GF-7 building dataset. The proposed KANBDNet outperforms all SOTA models, achieving ( F1-score, kappa coefficient, jaccard index, dice coefficient) of (87.75%, 75.50%, 78.87%, 80.69%) and (92.10%, 84.19%, 85.83%, 86.62%) on the NCVDL Cartosat-2S and GF-7 building datasets, respectively.
Sravya Nedungatt, Vibha Damodara Kevala, Shyam Lal et al.· Engineering Research Express· 0 citations
Accurate detection of land-use changes from aerial imagery is essential for urban development, environmental monitoring, and infrastructure management. While deep learning has advanced automated change detection, existing solutions remain sensitive to seasonal variations, lighting conditions, and image heterogeneity. This study presents an exploratory workflow-level evaluation of GPT-4o for semantic aerial imagery change detection and compares its behaviour with selected traditional, GIS-assisted, GAN-based, and U-Net-based approaches within the same imagery scenario. Unlike prior work focused on semantic segmentation, we investigate whether a general-purpose multimodal LLM can detect and describe changes without pixel-level training. NDVI exhibited inconsistent class separation; GAN-based mapping achieved a Structural Similarity Index Measure (SSIM) of 0.73 but lacked class fidelity; U-Net produced high accuracy for well-represented classes but struggled to generalize. GPT-4o achieved the best event-level performance, correctly identifying 89.17% of manually annotated changes and providing contextual descriptions and approximate spatial localization. Although promising, LLM performance depends on prompt specification and non-deterministic inference, raising reproducibility challenges. We address these by releasing prompt templates, raw outputs, and controlled inference settings. The results should therefore be interpreted as a case study of one proprietary multimodal model rather than as a comprehensive benchmark of all vision-language models. This exploratory study highlights the emerging potential of multimodal LLMs for interpretable and flexible geospatial analysis while outlining current limitations and future research directions.
Marek Ružička, Ján Staš, Stanislav Ondáš· IEEE Access· 0 citations
Land cover classification is a fundamental task in remote sensing with direct applications in environmental monitoring, urban planning, and disaster management. In this project, we design and implement a semantic segmentation pipeline using a modified U-Net architecture from scratch for pixel-level classification of satellite images. The network was trained on the DeepGlobe Land Cover Classification dataset with significant attention to model generalization and class imbalance through the integration of a custom composite loss function that combines Focal Tversky Loss and Weighted Categorical Crossentropy. Data augmentation techniques and per-class weighting were used to address class imbalance, especially in underrepresented classes such as barren land and rangeland. Postprocessing techniques like morphological operations and class-wise median filtering further refined the segmentation outputs. The model achieved a mean Intersection-over-Union (IoU) of 0.67, with notable performance across dominant land types. A full Streamlit-based web interface was developed to enable real-time visualization and interaction with the model’s predictions. Additionally, we initiated a fine-tuning phase targeting Nepal- specific satellite imagery to adapt the model to the country's unique geography. A major ongoing challenge involves curating a clean and labeled dataset of Nepalese satellite patches, which is critical for domain adaptation and achieving optimal segmentation performance on local data. The project demonstrates a full end-to-end pipeline from model design to web deployment and sets a foundation for scalable geospatial AI applications tailored for Nepal.
Yamraj Khadka, Rajat Pradhan· Journal of Engineering Issue...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
