运动估计
计算机视觉
人工智能
计算机科学
图像分辨率
过程(计算)
分辨率(逻辑)
亚像素分辨率
图像融合
运动(物理)
图像(数学)
图像处理
数字图像处理
操作系统
作者
Volodymyr Fedak,Zenoviy Veres,Adrian Nakonechny
出处
期刊:International Conference on Perspective Technologies and Methods in MEMS Design
日期:2010-04-20
卷期号:: 223-225
被引量:3
摘要
Super-resolution has relying on accurate motion estimation between the frames for a successful process. In this paper we propose to calculate quality coefficients for motion vectors and use calculated coefficients for accurate motion estimation process. All requirements for super resolution process: image registrations and fusion solutions are determined. Basic image spatial resolution algorithms were investigated. New enhanced super resolution algorithm was proposed. The proposed algorithm is designed to realize resolution upconversion on video sequences, trying to reduce the gap between high-resolution displays and low quality video content.
科研通智能强力驱动
Strongly Powered by AbleSci AI