节点文献
一种自适应视频图像超分辨率重构算法
Adaptive algorithm for super resolution reconstruction of video image
【摘要】 依据图像局部二阶统计量能够反映图像区域变化的特性,以参考帧的局部方差为参数建立凸集投影算子和区域修复阈值的条件,提出了一种自适应选取运动估计误差阈值和图像修复误差阈值的视频图像超分辨率重构方法。实验结果表明,与传统方法相比,该算法重构图像的主观质量得到了明显增强,提高了峰值信噪比。
【Abstract】 An adaptive algorithm for super resolution reconstruction of video image was proposed.Based on the character of the second order statistics parameter which could reflects image’s local variation,author adopted the local variance of refe ̄rence frame as the foundation to establish the operator for projection to convex set and local threshold value for pixel-repair.The experimental results show that the performance of this algorithm is better than the traditional method both in vision effect and peak signal to noise ratio.
【关键词】 超分辨率;
图像重构;
凸集投影;
运动估计误差;
边缘约束;
【Key words】 super resolution; image restoration; projection to convex set(POCS); motion estimation error; edge constraint;
【Key words】 super resolution; image restoration; projection to convex set(POCS); motion estimation error; edge constraint;
【基金】 国家自然科学基金资助项目(60272099);华为技术有限公司高校科技基金资助项目(YJCB2004007MU);华中科技大学研究生创新基金项目(JJ-05-071)
- 【文献出处】 计算机应用研究 ,Application Research of Computers , 编辑部邮箱 ,2007年08期
- 【分类号】TP391.41
- 【被引频次】1
- 【下载频次】239