Compressed sensing image reconstruction via adaptive sparse nonlocal regularization
Compressed sensing image reconstruction via adaptive sparse nonlocal regularization
复制标题
通过自适应稀疏非局部正则化压缩感知图像重建
DOI：
10.1007/s00371-016-1318-9
复制
发表时间：
2016-09
期刊：
影响因子：
3.5
通讯作者：
Shang Zhenhong
中科院分区：
文献类型：
--
作者：
Zha Zhiyuan;Liu Xin;Zhang Xinggan;Chen Yang;Tang Lan;Bai Yechao;Wang Qiong;Shang Zhenhong
压缩感知（CS）已被许多计算机视觉应用成功利用。然而，信号重建的任务仍然是具有挑战性的，特别是当我们只有图像的CS测量（CS图像重建）。与传统图像恢复的任务（例如，图像去噪、去模糊和修补等），CS图像重建具有部分结构或局部特征。很难从CS图像本身建立用于CS图像重建的字典。很少有研究显示出有希望的重建性能，因为大多数现有方法采用固定的基集（例如，小波、DCT和梯度空间）作为字典，缺乏对图像局部结构的适应性。在本文中，我们提出了一种自适应稀疏非局部正则化（ASNR）的CS图像重建方法。在ASNR中，一个有效的自适应学习字典被用来大大减少伪影和细节的损失。字典是紧凑的，从重建图像本身而不是自然图像数据集学习。此外，图像稀疏非局部（或非局部自相似）的先验集成到正则化项，因此ASNR可以有效地提高CS图像重建的质量。为了提高ASNR的计算效率，分裂Bregman迭代为基础的技术也被开发，它可以表现出更好的收敛性能比迭代收缩/阈值方法。大量的实验结果表明，所提出的ASNR方法可以有效地重建精细结构和抑制视觉伪影，优于国家的最先进的性能方面的PSNR和视觉测量。
Compressed sensing (CS) has been successfully utilized by many computer vision applications. However,the task of signal reconstruction is still challenging, especially when we only have the CS measurements of an image (CS image reconstruction). Compared with the task of traditional image restoration (e.g., image denosing, debluring and inpainting, etc.), CS image reconstruction has partly structure or local features. It is difficult to build a dictionary for CS image reconstruction from itself. Few studies have shown promising reconstruction performance since most of the existing methods employed a fixed set of bases (e.g., wavelets, DCT, and gradient spaces) as the dictionary, which lack the adaptivity to fit image local structures. In this paper, we propose an adaptive sparse nonlocal regularization (ASNR) approach for CS image reconstruction. In ASNR, an effective self-adaptive learning dictionary is used to greatly reduce artifacts and the loss of fine details. The dictionary is compact and learned from the reconstructed image itself rather than natural image dataset. Furthermore, the image sparse nonlocal (or nonlocal self-similarity) priors are integrated into the regularization term, thus ASNR can effectively enhance the quality of the CS image reconstruction. To improve the computational efficiency of the ASNR, the split Bregman iteration based technique is also developed, which can exhibit better convergence performance than iterative shrinkage/thresholding method. Extensive experimental results demonstrate that the proposed ASNR method can effectively reconstruct fine structures and suppress visual artifacts, outperforming state-of-the-art performance in terms of both the PSNR and visual measurements.
登录
查看更多内容
通过自适应稀疏域选择和自适应正则化进行图像去模糊和超分辨率
影响因子：
10.6
作者：
Dong, Weisheng;Zhang, Lei;Wu, Xiaolin
通讯作者：
Wu, Xiaolin
学习汇集高级特征以进行人脸表示
影响因子：
3.5
作者：
Huang Renjie;Ye Mao;Xu Pei;Li Tao;Dou Yumin
通讯作者：
Dou Yumin
DOI：
10.1017/cbo9780511794308
发表时间：
2012
期刊：
--
影响因子：
--
作者：
Gitta Kutyniok
通讯作者：
Gitta Kutyniok
DOI：
10.1007/s00371-007-0139-2
发表时间：
2007-08
期刊：
The Visual Computer
影响因子：
--
作者：
Ubiratã A. Ignácio;C. Jung
通讯作者：
Ubiratã A. Ignácio;C. Jung
DOI：
10.1109/cvpr.2012.6247930
发表时间：
2012-06
期刊：
2012 IEEE Conference on Computer Vision and Pattern Recognition
影响因子：
--
作者：
Shenlong Wang;Lei Zhang;Yan Liang;Q. Pan
通讯作者：
Shenlong Wang;Lei Zhang;Yan Liang;Q. Pan