This book will cover basic properties, features and models for very specific Earth observation (EO) cases recorded by very high-resolution (VHR) multispectral, hyperspectral, Synthetic Aperture Radar (SAR), and multi-temporal observations. It will cover pre-processing methods applied on satellite images and various deep learning techniques for various applications such as identifying land cover features, object detection, cloud detection, crop classification, oil spill detection, target recognition , land subsidence , etc . We will look for the solutions to build pre-trained deep neural networks for: (A) various earth resource applications, (B) pre-processing techniques and (C) annotating objects on earth surface. Well-labelled training images are still key to the quality of any algorithms built upon satellite imagery. Annotating any areas on earth requires a good image interpretation and field visit. We will discuss several automated methods for spatial annotation. Thus, compiling existing knowledge with direct applicability of the technology will open new avenues in this area.