A Review of Vision SLAM-based Closed-loop Inspection

As unmanned technology and virtual reality have developed rapidly, visual simultaneous localization and map building (SLAM) has become an important research area. Scene description is one of the most challenging problems in the field of visual SLAM. Creating image descriptors is one of the most fundamental problems in scene recognition. During long-term use, these descriptors must be insensitive to changes in illumination and viewpoint. The purpose of this paper is to give a brief introduction to visual SLAM closed-loop detection and to summarize existing solutions for scene recognition in two directions, traditional methods and deep learning, as well as to analyze the advantages and disadvantages of these methods. Finally, we look forward to the directions that can help drive the future development of closed-loop detection scene description technology.