| 图像自适应分块单幅超分辨率算法 |
| |
| 引用本文： | 李展,,吴少春,,彭青玉.图像自适应分块单幅超分辨率算法[J].系统工程与电子技术,2015,37(10):2411-2417. |
| |
| 作者姓名： | 李展 吴少春 彭青玉 |
| |
| 作者单位： | (1. 暨南大学计算机科学系, 广东 广州 510632; 2. 深圳市超节点网络科技有限公司,;
广东 深圳 518100; 3. 暨南大学天体测量、动力学和空间科学研究;
中法联合实验室, 广东 广州 510632) |
| |
| 摘 要： | 在单幅超分辨率图像重建算法中,基于最大后验估计(maximum a posteriori,MAP)算法重建效果和抗噪性能较好,但时空复杂度较高。为了提高模板卷积MAP(template convolution based MAP,TC-MAP)算法的运行效率,降低内存消耗,提出了基于图像内容的自适应分块TC-MAP新算法,研究了图像分块的最佳尺寸,并根据子块图像的平均梯度,对平滑区域的多个子块进行合并降低分块边界效应的影响,同时采用边界延长进一步抑制分块效应。实验结果表明,算法有效减少了TC-MAP算法的运行时间和内存开销,同时保持重建图像质量与原TC-MAP算法差别不大。
|
Image adaptive block single frame super resolution algorithm |
| |
| LI Zhan,,WU Shao Chun,,PENG Qing Yu,.Image adaptive block single frame super resolution algorithm[J].System Engineering and Electronics,2015,37(10):2411-2417. |
| |
| Authors: | LI Zhan WU Shao Chun PENG Qing Yu |
| |
| Institution: | (1. Department of Computer Science, Jinan University, Guangzhou 510632, China; 2. Supernode Network;
Technology Corporation Limited, Shenzhen 518100, China; 3. Sino France Joint Laboratory for;
Astrometry, Dynamics and Space Science, Jinan University, Guangzhou 510632, China) |
| |
| Abstract: | Single frame super resolution (SR) image reconstruction algorithms based on maximum a posteriori (MAP) have preferable reconstruction results and noise robustness but high time and space complexity. To improve efficiency of template convolution based MAP (TC MAP) algorithm and decrease its memory consumption, a new image content based adaptive block TC MAP algorithm is proposed. Optimal size of image block is studied, and according to average gradients of image blocks, several blocks in smooth areas are combined to reduce the influence on block boundary effects, further block boundary extension is used to suppress block effects. Experiments show that both run time and memory consumption of TC MAP algorithms are decreased, while the quality of reconstructed images makes little difference. |
| |
| Keywords: | |
|
| 点击此处可从《系统工程与电子技术》浏览原始摘要信息 |
| 点击此处可从《系统工程与电子技术》下载免费的PDF全文 |
|
