摘要
为了改进图像恢复效果,减少运行时间,提出了一种基于改进的多层小波分解压缩感知图像的处理方法。该方法利用小波变换对图像进行多层分解,根据小波域高频系数分布特点通过高斯矩阵随机观测,利用正交匹配追踪算法（OMP）恢复高频系数,最后通过小波逆变换重构图像。实验结果表明,本文的算法优于传统算法,峰值信噪比（PSNR）平均提高了4~6 d B,运行时间缩短了1~2个量级,为物联网、无线传输技术提供了更好的可能性。
In order to improve the effect of image restoration and reduce the operation time, a compressed sensing method based on improved multi-layer wavelet decomposition is proposed. The image is divided into multi-blocks by multilayer wavelet transform and the Gaussian matrix is used to observe the high frequency based on the characteristics in domain. For the reconstruction, high frequency can be recovered by the matching pursuit algorithm （OMP） and the image can be reconstructed by the inverse wavelet transformation. Simulation results demonstrate that the proposed algorithm is better than the traditional algorithm, and the PSNR is improved about 4 ~ 6 dB, and the operating time is shortened about 1 - 2 magnitude. It provides a better possibility for Internet of things and wireless transmission technology.
出处
《山西电子技术》
2016年第6期5-6,25,共3页
Shanxi Electronic Technology
基金
国家自然科学基金(31170668)
关键词
压缩感知
多层
分块
小波变换
高斯矩阵
正交匹配追踪算法
重构
compressed sensing
multi layer
blocks
wavelet transform
Gaussian matrix
matching pursuit algorithm
recon- struction
