부산대학교 도서관

With multiple applications and use cases of Digital Elevation Models (DEMs), focusing significantly on forming 3D models. It is used in various research areas such as flood modeling, agriculture, satellite navigation, farming, and forestry. This research paper presents a methodology for extracting Digital Elevation Models (DEMs) from satellite images using U-Net. The U-NET is a Deep Learning Model architecture that is commonly used for image segmentation tasks. It is a fully convolutional neural network that uses an encoder-decoder architecture to extract features from the input image and generate a pixel wise output. The U-Net model is ideal for DEM extraction from satellite images as it can effectively distinguish between the different features present in the images and accurately segment the ground elevation ranges. The creation of the dataset was a crucial step in the project as it ensured that the U - N et model was trained on a diverse set of images. The SWISSIMAGE dataset provided high-resolution RGB imagery, while the SRTM dataset provided the corresponding mask images with ground elevation ranges. The results obtained from the project indicate that the model can perform reasonably well even with limited data and without any significant pre-processing of the input dataset. However, there are still some challenges and limitations corresponding to the use of U-Net for DEM extraction. In conclusion, this study provides a valuable contribution to the field of remote sensing and computer vision, highlighting the potential of U-Net for DEM extraction from satellite images. In addition, the latest improvements in both software and hardware have made it possible to carry the processing with remarkable speed, even on moderately powerful hardware. We anticipate that our research will act as a catalyst for further advancements and will facilitate the integration of this research in generating, updating, and analysis of DEMs. However, further research is needed to address the limitations and challenges associated with this approach and to develop more accurate and reliable models for DEM extraction