Moving Object Detection Based on Flexible Edge Matching

Moving object detection is an important branch and foundation of computer vision; it has extensive application prospects in many fields, such as traffic, military application, industries, and bio-medical, et al., and becomes a hot research topic in computer vision field. Because of its inherent complexity, moving object detection still faces lots of challenges. In this paper, based on the existing research achievements, the methods of moving object detection in dynamic environment are studied deeply, a knowledge-based flexible edge matching algorithm is put forward, and the effectiveness of the proposed matching algorithm in moving object detection is also demonstrated. The research results here can be provided as the reference for target detection and tracking and some other applications.