In Vision Simultaneous Localization and Mapping (VSLAM), feature detection is used in landmark extraction and data association. It examines each pixel to find interesting part of an image that would differentiate the landmark and the less important image details. There are numerous studies in this field but they are scattered in many journals and proceedings which would require many hours just to find related material. Therefore, this research has grouped important studies done in this field to be analyzed by future researcher. Feature detection techniques such as Harris, Scale Invariant Feature Transform (SIFT), Speeded-Up Robust Features (SURF), Features from Accelerated Segment Test (FAST) and etc. is discussed in this study. A background history of each technique, their evolution and performance comparison is presented.
展开▼
