- 期刊
Construction of a Panoramic Image for a Video-Based Tracking System
並列摘要
Video surveillance and monitoring (VSAM) has been a popular research topic in recent years. Many researchers have focused their studies on image processing, pattern recognition, or artificial intelligence for object monitoring or event detection. In this study, an immersed video-based monitoring environment was developed by video processing and spherical mosaicking techniques. Image sequences were grabbed to construct the panoramic background images from a program controlled pan-tilt-zoom (PTZ) camera. These images were stitched by spherical mosaicking techniques. In addition, the moving targets were detected, tracked and segmented from video sequences. The PTZ camera was triggered by the predicted data to continuously capture the object images when an object was out of the field of view (FOV). Two experimental environments, an indoor and an outdoor, were constructed to show the validity and effectiveness of the proposed approach.