作者: Mahmood Amintoosi , Mahmood Fathy , Nasser Mozayani
DOI:
关键词:
摘要: Moving objects is one of the main bottlenecks in many computer vision problems including video super resolution. Video Super-resolution algorithms reconstruct a high resolution from low video. Recent advances Super-Resolution techniques show trends towards model-based or example-based approaches. In this article new method for super-resolution problem based on our recent paper proposed with at same time handles moving efficiently. The overall hypothesis, which some high-resolution(HR) images scene are available. Instead previous example methods, do blocking, here whole training image mapped to each frame. Proper mapping HR onto LR frames has been done using combination feature-based and area-based registration methods. After creating suitable mask handling objects, frame fused registered image. simulation results better performance algorithm compare other superresolution