Building detection methods from remotely sensed images

With the availability of high-resolution satellite imagery, new applications have been developed for solving geospatial issues in urban regions. Building detection from remote sensing images has been an active area of research due to its broad range of applications, including city modelling, map updating and urban monitoring. The manual processing of an image is a time-consuming and laborious task. Therefore, researchers have developed methods that involve less or no human effort. At present, building detection has improved through various automated and semi-automated methods/algorithms/ techniques suggested in various studies. The objective of the present study is to review the efforts of such studies. Here, the building detection methods are categorized into six groups: (i) low-level feature-based methods, (ii) snake models, (iii) graph-based methods, (iv) shadow detection-based methods, (v) cognition-based methods and (vi) deep learning models. We hope that this study will aid the researchers working in this domain.