基于压缩多信号分类算法的森林区域极化SAR层析成像

该文研究了一种基于压缩多信号分类算法的森林区域极化SAR层析成像方法。其具体步骤包括:全极化的SAR接收成像区域的反射回波,利用各极化通道的信号建立多观测向量模型;应用小波基对高程向结构进行稀疏表示,采用压缩多信号分类算法对观测区域的高程向后向散射系数进行重建,实现对森林区域层析成像。最后,通过仿真实验、PolSARpro仿真数据和德宇航E-SAR的P-波段数据验证了该方法在同等测量精度的要求下可以有效减少SAR层析成像所需的航过数,同时降低了虚假目标的出现概率。     

基于压缩感知的鲁棒可分离的密文域水印算法

为了满足密文域水印嵌入的需要,该文基于压缩感知技术,提出一种鲁棒可分离的密文域水印算法。首先,内容拥有者将图像进行不重叠分块,利用边缘检测手段划分重要块和非重要块。重要块用传统加密方式进行加密,非重要块用压缩感知技术进行加密,同时为水印嵌入留出一定空间,然后根据嵌入密钥,实现二值水印的密文嵌入。在接收端获取图像内容和水印的方式是可分离的,同时根据含水印的密文图像块的像素分布特性可重新判断块的属性,避免了传输块属性信息。此外,水印信息重复4次嵌入在密文图像的不同区域,保证了水印的鲁棒性。实验结果显示所提方案在抵抗适度攻击时具有鲁棒性和安全性。     

Compressive optical image watermarking using joint Fresnel transform correlator architecture

A new optical image watermarking technique based on compressive sensing using joint Fresnel transform correlator architecture has been presented. A secret scene or image is first embedded into a host image to perform optical image watermarking by use of joint Fresnel transform correlator architecture. Then, the watermarked image is compressed to much smaller signal data using single-pixel compressive holographic imaging in optical domain. At the received terminal, the watermarked image is reconstructed well via compressive sensing theory and a specified holographic reconstruction algorithm. The preliminary numerical simulations show that it is effective and suitable for optical image security transmission in the coming absolutely optical network for the reason of the completely optical implementation and largely decreased holograms data volume.     

压缩感知SAR成像中的运动补偿

由于其具有压缩采样特性,压缩感知在高分辨SAR成像技术中得到了广泛应用。然而作为一种参数化的成像方法,基于压缩感知的成像方法对位置误差非常敏感。位置误差会造成图像偏离真实位置、散焦、甚至根本不能成像。该文针对SAR压缩成像系统中存在的运动误差,分析了平台非理想运动对回波信号的调制机理和运动相位误差对信号稀疏表征的影响,提出了基于传感器测量数据进行运动补偿的压缩感知SAR成像方法,通过在稀疏矩阵中引入附加项完成空不变运动误差的补偿。该方法不仅能以少量的测量孔径和测量数据获得重建目标空间的足够信息而且能有效降低运动误差对成像质量的影响,实现高分辨成像。     

WMSN中基于VSQI压缩反馈的协作节点选择

无线传感器网络中采用协同通信技术进行收集、交换和分析数据,可以有效提高信息精确度,并大大减少节点的能耗。其中协作节点的选择是一个重要问题。考虑在WMSN中应用信息压缩反馈技术,将分布在临近地理位置上的传感器节点的视频流质量指标（Video Stream Quality Index, VSQI）通过控制信道反馈给汇聚节点,可以为协同节点的选择提供支持。本文根据人类视觉特征,提出了一种用于视频流质量估计的方法,并据此综合考虑VSQI信号的设计。由于WMSN中巨大的信息反馈量,将压缩感知（Compressive Sensing, CS）理论应用到VSQI压缩反馈中进行了研究,寻找并验证了与VSQI信号相关的随机测量矩阵与重建准则,实验结果证明基于压缩感知的压缩反馈方法可以降低感知节点端处理复杂度、增强反馈压缩程度,从而确定WMSN中传输节点的协作节点。      

The Effect of Strain Rate,Specimen Geometry and Lubrication on Responses of Aluminium AA2024 in Uniaxial Compression Experiments

The ability to observe and quantify intrinsic material response to loading at different rates of strain has been improved by reducing the errors of mechanical characterisation in uniaxial compression experiments. In order to perform comparisons of the results from uniaxial compression tests used to characterise mechanical properties of aluminium alloys at different strain rates, it is necessary to reduce errors resulting from factors such as specimen design. In this study, the effects of strain rate, specimen geometry and lubrication on the compressive properties of aluminium AA2024 alloy were quantitatively investigated by measuring the mechanical behaviour of this alloy as functions of strain rate, specimen aspect ratio and lubrication condition. Both the deformation history and the failure mode were identified using low and ultrahigh speed photography. The interaction of factors influencing the measured stress-strain response was quantified, and suitable specimen aspect ratios for compression tests at different strain rates were identified.     

Kink Band Formation in Wood Species Under Compressive Loading

The design of wood structures, such as columns and beams, requires the knowledge of the compressive behavior of wood. Several wood species, hardwoods and softwoods, are tested under compression parallel to grain to investigate experimentally the formation of kink bands and the relationship between this localized mode of failure and wood anatomy. In order to determine the correct test method for successful observation of kink banding, an experimental plan is first carried out. Then, several characteristics of kink bands, as well as the compressive behavior of the specimens, are comparatively analyzed for the various wood species. Based on these observations and some theoretical predictions corroborated by experimental data, it is shown that compressive kinking strength of wood is governed mainly by its yield strength in shear and by certain features of its anatomy related to the so-called ray cells.     

从稀疏到结构化稀疏:贝叶斯方法

稀疏分解算法是稀疏表达理论和压缩感知理论中的核心问题,也是当前信号处理领域的一个热门话题。近年来,研究人员发现除了稀疏以外,如果引入稀疏系数之间的相关性先验信息,可以大大提高稀疏分解算法的精度,这种方法称为“结构化稀疏分解算法”。本文归纳和总结了从稀疏到结构化稀疏的信号模型,并且介绍了两种不同的贝叶斯稀疏（或者结构化稀疏）算法,以及从稀疏到结构化稀疏贝叶斯稀疏分解算法的扩展。同时,本文还介绍了结构化稀疏分解算法在医学信号处理和语音信号处理中的应用。      

基于压缩感知理论的多普勒解模糊处理

多目标的多普勒解模糊处理是低脉冲重复频率(PRF)雷达的关键技术之一,论文提出一种新的基于压缩感知(CS)理论的多普勒解模糊处理方法,利用多重PRF方式下相参处理间隔内的时域欠采样特性及多普勒频谱的稀疏特性,构造了多普勒解模糊的CS模型,采用正交匹配追踪(OMP)算法直接估计出无模糊多普勒谱的幅度响应,可实现PRF分组参差方式下对多个目标的解多普勒模糊处理,仿真结果验证了算法的有效性。     

基于感知矩阵统计相关系数最小化的压缩感知雷达波形优化设计

为了改善压缩感知雷达(Compressive Sensing Radar, CSR)目标参数提取的性能,该文提出一种最小化感知矩阵统计相关系数的CSR波形优化设计方法。文中首先建立了通用的CSR系统模型,推导了最小化感知矩阵统计相关系数的波形优化目标函数,其次以多相编码信号作为优化码型并采用遗传算法对目标函数进行优化求解。优化设计的波形使得感知矩阵子矩阵近似正交程度达到最优,与传统波形相比,能够有效降低目标参数估计误差,提高可检测目标个数的上限,改善了CSR目标参数提取的准确性和鲁棒性。计算机仿真验证了该方法的有效性。