阈值法在毫米波目标辐射信号去噪中的应用研究

小波域阈值法去噪以其效果好,易编程实现而广泛应用到图像及信号的去噪中。该文在分析了毫米波目标辐射信号的小波系数特征后,提出使用非负小波系数代替信号的小波系数。对于确定的阈值,推导了重构信号均方差最小时,非负小波系数的去噪方法,实验表明该文算法具有较好的去噪效果。     

Joint image-to-image translation with denoising using enhanced generative adversarial networks

Impressive progress has been made recently in image-to-image translation using generative adversarial networks (GANs). However, existing methods often fail in translating source images with noise to target domain. To address this problem, we joint image-to-image translation with image denoising and propose an enhanced generative adversarial network (EGAN). In particular, built upon pix2pix, we introduce residual blocks in the generator network to capture deeper multi-level information between source and target image distribution. Moreover, a perceptual loss is proposed to enhance the performance of image-to-image translation. As demonstrated through extensive experiments, our proposed EGAN can alleviate effects of noise in source images, and outperform other state-of-the-art methods significantly. Furthermore, we experimentally indicate that the proposed EGAN is also effective when applied to image denoising.     

差分投影快速定向图像插值算法的优化

针对图像放大时分辨率会降低,图像的感知质量变差等问题,并考虑到计算的复杂度,提出了一种差分投影快速定向图像插值算法。该方法通过对低分辨率图像的梯度扩散确定高分辨图像的丢失像素,由特殊的去噪方法对该算法进行优化,并与最近邻插值法与双线性插值算法等相比较。实验结果表明,该方法具有较高的分辨率和较低的计算复杂度,优于其他方法。     

Nonparametric Bayesian dictionary learning algorithm based on structural similarity

Though nonparametric Bayesian methods possesses significant superiority with respect to traditional comprehensive dictionary learning methods,there is room for improvement of this method as it needs more consideration over the structural similarity and variability of images.To solve this problem,a nonparametric Bayesian dictionary learning algorithm based on structural similarity was proposed.The algorithm improved the structural representing ability of dictionaries by clustering images according to their non-local structural similarity and introducing block structure into sparse representing of images.Denoising and compressed sensing experiments showed that the proposed algorithm performs better than several current popular unsupervised dictionary learning algorithms.     

双Haar小波变换系数的MAP估计及在图像去噪中的应用

小波变换作为一种新的工具,在信号去噪中得到了重要的应用。本文对双Haar小波变换系数,提出了MAP的估计方法,并对其在图像去噪中的应用进行了讨论。实验表明所提出的小波收缩算法与软门限方法相比较,用于图像去噪时可以给出更好的结果。     

Dynamic denoising studies in wideband radio channel measurement and modeling

The fixed level and dynamic denoising method was studied based on indoor-to-outdoor measured channel impulse responses ORs） at 5.25 GHz with radio frequency （RF） 100 MHz bandwidth. It is found that the dynamic ranges, peak powers and noise floors of the IRs are with close correlations. The comparisons with different denoising methods are given by deriving the power delay profiles （PDPs）, root mean square （RMS） delay spread （RMS DS）, number of paths （NOPs） and Ricean K-factors. It is concluded that the traditional fixed level noise cut is under estimate of DS and NOPs. The Ricean K-factors are of little sensitive to noise cut irrespective of what kind of methods applied. The PDPs are not very sensitive to the fixed level noise cut, however, obvious changes can be found by dynamic noise cut. The dynamic noise cut is preferred when clear noise floors is observed and decided from the measured IRs, it＇s of importance in data post processing for wideband radio channel measurements as well as the relevant modeling work.     

非高斯双变量模型contourlet图像去噪

Contourlet变换是继小波变换之后的又一新变换.由于contourlet变换的多尺度和多方向特性,能有效地捕获到自然图像中的轮廓,并对其进行稀疏表示.详细分析了图像contourlet系数的统计特性,并利用非高斯双变量分布对系数层间相关性进行建模.最后,将此分布应用于图像去噪,就PSNR、NMSE和视觉质量这三方面的评价指标与contourlet HMT和小波阈值法进行了比较.实验结果表明：算法能获得较好的结果,尤其是对于含有丰富纹理的图像.     

基于线阵CCD与小波去噪的光谱数据采集系统研究

为实现光谱数据高效稳定的采集、存储、传输及显示,提出了基于线阵CCD与小波去噪的光谱数据采集系统的实现方法.系统完全依靠FPGA控制线阵CCD模块及AD转化模块、信号调理模块、电源模块,完成光谱数据采集、存储.通过USB2.0接口将数据稳定传输至上位机,对光谱数据进行中值与小波混合算法去噪处理,最终获取光谱曲线.系统使...     

基于小波变换的边缘保留图像去噪改进算法

针对在图像去噪过程中,如何有效地保留图像边缘等重要特征信息的问题,提出了一种基于小波变换的图像去噪改进算法。对图像进行多尺度小波分解,将各子带小波系数进行自适应阈值化处理,边缘成分的阈值由子带阈值和给定的相关权重计算得到,从而有效保留图像边缘信息。分别对Tracy和Building图像进行处理,实验结果表明,与BayesShrink等4种传统方法相比较,改进算法不仅可以有效去除不同程度的加性高斯白噪声,很好地保留图像边缘等重要特征信息,而且具有较高的峰值信噪比。     

基于模极大值的宽带雷达信号去噪方法

经典Mallat小波模极大值降噪法因其程序复杂度高、计算量大难以满足宽带雷达信号处理的性能要求.在模极大值降噪算法的基本思想上,提出基于“路径”的模极大值降噪算法,并利用改进的自适应阀值估计对模极大值序列进行预处理及利用分段多项式插值对模极大值序列进行快速重构.选用LFM信号为宽带雷达发射信号,对目标回波高分辨率像进行模极大值降噪处理.Maltab仿真结果表明,改进后的模极大值降噪算法具有良好的去噪性能,有效提高宽带雷达回波信号的信噪比,且能保留目标大部分特征信息.