• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>International Conference on Cloud Computing and Security >Adaptive Hybrid Wavelet Regularization Method for Compressive Imaging
【24h】

Adaptive Hybrid Wavelet Regularization Method for Compressive Imaging

机译：自适应混合小波正则化压缩成像方法

获取原文
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

This paper proposes a hybrid method that simultaneously considers sparsity in wavelet domain and image self-similarity by using wavelet LI norm, nonlocal wavelet L0 norm regularization in image compressive sensing (CS) recovery. An auxiliary variable is then introduced to decompose this composite constraint problem into two simpler regularization sub-problems. Based on Fast Iterative Shrinkage-Thresholding Algorithm (FISTA), the sub-problems corresponding to the wavelet LI norm and the nonlocal wavelet L0 norm are then solved by soft thresholding and adaptive hard thresholding respectively. The threshold of the later is decreased according to the energy of measurement error, leading to an adaptive hybrid regularization method. Experimental results show that it outperforms several excellent CS techniques.
机译：本文提出了一种混合方法，该方法同时使用小波LI范数，非局部小波L0范数正则化来同时考虑小波域的稀疏性和图像自相似性，以进行图像压缩感知（CS）恢复。然后引入辅助变量，以将该复合约束问题分解为两个更简单的正则化子问题。基于快速迭代收缩阈值算法（FISTA），分别通过软阈值和自适应硬阈值分别求解与小波LI范数和非局部小波L0范数相对应的子问题。后者的阈值根据测量误差的能量而减小，从而导致自适应混合正则化方法。实验结果表明，它优于几种出色的CS技术。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号