Automatic and unsupervised detection and extraction of objects from satellite images has been an important research topic in the field of photogrammetry and remote sensing in recent years. Several commercial satellites with high-resolution imaging capability and which can provide multispectral images were launched into different orbits around Earth. High resolution satellite imagery is being increasingly employed for large-scale topographic mapping, and especially for updating Geographic Information System (GIS) databases. Satellite images have inhomogeneous properties that make it hard to develop generic algorithms for object detection. The image quality may vary widely depending on resolution, sensor type, sun elevation and azimuth angles. This book analyzes structural feature extraction from satellite images which exclusively discusses algorithms for roads, buildings and bridges. Buildings may have complicated non-linear structures and can be occluded by other buildings or trees. So, as a supportive tool, detection and elimination of shadows and clouds are also considered. The algorithms developed are simulated and competitive results are obtained for the methods.