结合压缩感知和曲波的天文图像去噪
发布时间：2018-10-31 16:09
【摘要】：在天文图像去噪中,为了提高迭代曲波阈值算法的去噪重建性能,提出了基于循环平移和曲波维纳滤波的压缩感知迭代重构算法。首先,使用基于曲波阈值的循环平移方法对重构图像进行调整以抑制重构图像中的伪吉布斯效应;接着,用提出的曲波维纳滤波算子替代小波阈值在迭代过程中对图像曲波系数进行筛选以进一步提高重构图像的质量。通过对添加高斯白噪声的Lena图像和月球图像进行重构实验,分析本文算法和当前主流算法的性能。实验结果表明,与传统的压缩感知迭代曲波阈值算法相比,本文算法能够获得较优的去噪性能,有效地保护天文图像的细节信息,峰值信噪比大约提高了2.6~3.2dB。
[Abstract]:In order to improve the performance of iterative Qu Bo threshold algorithm in astronomical image denoising, a compression sensing iterative reconstruction algorithm based on cyclic translation and Qu Bo Wiener filter is proposed. Firstly, the cyclic translation method based on Qu Bo threshold is used to adjust the reconstructed image to suppress the pseudo-Gibbs effect in the reconstructed image. Then, the proposed Qu Bo Wiener filter operator is used instead of the wavelet threshold in the iterative process to filter the Qu Bo coefficient of the image to further improve the quality of the reconstructed image. Based on the experiments of Lena and moon images with Gao Si white noise, the performance of this algorithm and the current mainstream algorithms are analyzed. The experimental results show that compared with the traditional compression sensing iterative Qu Bo threshold algorithm, the proposed algorithm can obtain better denoising performance and effectively protect the detailed information of astronomical images.
【作者单位】： 哈尔滨工业大学控制与仿真中心;
【基金】：国家自然科学基金资助项目(No.61074127)
【分类号】：TP391.41
[Abstract]:In order to improve the performance of iterative Qu Bo threshold algorithm in astronomical image denoising, a compression sensing iterative reconstruction algorithm based on cyclic translation and Qu Bo Wiener filter is proposed. Firstly, the cyclic translation method based on Qu Bo threshold is used to adjust the reconstructed image to suppress the pseudo-Gibbs effect in the reconstructed image. Then, the proposed Qu Bo Wiener filter operator is used instead of the wavelet threshold in the iterative process to filter the Qu Bo coefficient of the image to further improve the quality of the reconstructed image. Based on the experiments of Lena and moon images with Gao Si white noise, the performance of this algorithm and the current mainstream algorithms are analyzed. The experimental results show that compared with the traditional compression sensing iterative Qu Bo threshold algorithm, the proposed algorithm can obtain better denoising performance and effectively protect the detailed information of astronomical images.
【作者单位】： 哈尔滨工业大学控制与仿真中心;
【基金】：国家自然科学基金资助项目(No.61074127)
【分类号】：TP391.41
【相似文献】
相关期刊论文 前10条
1 王化U，
本文编号：2302817
本文链接：https://www.wllwen.com/kejilunwen/ruanjiangongchenglunwen/2302817.html
