基于正则化处理的超分辨率重建

针对超分辨率重建的病态反问题进行研究。该文首先介绍超分辨率重建的数学模型,分析了最小二乘估计及其病态性。其次给出病态问题的正则化泛函,并提出一种自适应动态确定正则化系数的方法,主要研究了迭代算法的收敛性和参数选择等问题。实验结果验证了方法的有效性。     

Multiframe super-resolution based on half-quadratic prior with artifacts suppress

The multiframe super-resolution (SR) technique aims to obtain a high-resolution (HR) image by using a set of observed low-resolution (LR) images. In the reconstruction process, artifacts may be possibly produced due to the noise, especially in presence of stronger noise. In order to suppress artifacts while preserving discontinuities of images, in this paper a multiframe SR method is proposed by involving the reconstruction properties of the half-quadratic prior model together with the quadratic prior model using a convex combination. Moreover, by analyzing local features of the underlined HR image, these two prior models are combined by using an automatically calculated weight function, making both smooth and discontinuous pixels handled properly. A variational Bayesian inference (VBF) based algorithm is designed to efficiently and effectively seek the solution of the proposed method. With the VBF framework, motion parameters and hyper-parameters are all determined automatically, leading to an unsupervised SR method. The efficiency of the hybrid prior model is demonstrated theoretically and practically, which shows that our SR method can obtain better results from LR images even with stronger noise. Extensive experiments on several visual data have demonstrated the efficacy and superior performance of the proposed algorithm, which can not only preserve image details but also suppress artifacts.     

一种稳健的多视频时空超分辨率重建算法

为了提高视频序列的时间分辨率和空间分辨率,该文提出了一种基于一阶范数和时空总变分正则化方法的时空超分辨率视频重建算法。该算法利用同一场景的多个具有子像素空间位移偏差和子帧率时间偏差的低分辨率视频序列,重建得到一个高时空分辨率视频序列。在求解过程中不需要直接构造大型矩阵,大大降低了对内存的要求。实验结果表明该算法是有效的,且对成像模型估计误差具有一定的鲁棒性。     

Principles of minimum variance robust adaptive beamforming design

Robustness is typically understood as an ability of adaptive beamforming algorithm to achieve high performance in the situations with imperfect, incomplete, or erroneous knowledge about the source, propagation media, and antenna array. It is also desired to achieve high performance with as little as possible prior information. In the last decade, several fruitful principles to minimum variance distortionless response (MVDR) robust adaptive beamforming (RAB) design have been developed and successfully applied to solve a number of problems in a wide range of applications. Such principles of MVDR RAB design are summarized here in a single paper. Prof. Gershman has actively participated in the development and applications of a number of such MVDR RAB design principles.     

基于非局部全变分正则化优化的单幅雾天图像恢复新方法

该文针对无雾图像具有高灰度对比度且大气遮罩局部平滑的特性,提出一种基于非局部全变分正则化优化的单幅雾天图像恢复新方法。先构建一种基于非局部全变分正则化的有约束优化算法对大气遮罩进行估计,然后通过优化Bregman分离迭代法求解非局部Rudin-Osher-Fatemi模型获得准确的大气遮罩,进而从雾天场景图像恢复出场景图像。实验结果表明,所提新方法可以有效地对雾天降质图像进行复原,对多纹理复杂区域的恢复效果也较好。     

分布式认知无线网络中的频谱接入凸规划算法

针对分布式认知无线网络中所可能存在的多频道隐终端问题,以及分布式协调方法(DCF)在随机冲突下性能下降的问题,该文以最优化网络的效用值为目标,将各个认知节点的接入概率作为求解对象,在分布式认知无线网络中建立了一个通用的凸规划模型。同时,基于凸规划算法,提出了一种分布式的信道接入 (DSAC) 算法对目标问题进行求解,从而决定认知网络中每个节点以及每条链路的接入概率。仿真结果表明,DSAC算法能够在较少的迭代次数内实现收敛;与普通的DCF方法相比,DSAC算法可以有效地提高网络中同时传输的链路个数,提高了频谱利用率以及网络吞吐量。     

线性约束条件下任意凸函数的神经网络优化模型

该文提出了线性约束条件下任意凸函数的神经网络优化模型,所构造的能量函数的平衡点即为原问题的最优解,克服了传统的神经网络优化方法所存在的问题,网络是全局稳定的,并能收敛到最优点,计算机仿真结果证明了本文方法的有效性。     

一种全局最小化的图像分割方法

曲线活动模型是图像分割中应用广泛且成功的一类模型,但由于能量泛函的非凸构造,使得其分割结果往往陷入局部解的困境。为了克服这一点,该文在已有的曲线活动模型之一背景去除模型之上,从Heaviside函数的近似入手,提出了凸的能量泛函,并对其最小化,得到了相应的全局最小解求解方程。实验表明,该方法分割结果准确,分割速度快,具有一定的抗噪性,且对初始曲线的位置选取无特殊要求。     

基于实值处理的联合波束域双基地MIMO雷达测角算法

该文提出一种基于实值处理的联合波束域双基地多输入多输出(MIMO)雷达测角方法。与直接采用离散傅里叶变换(DFT)空域滤波器的传统波束域方法不同,该方法的空域滤波器通过凸优化方法进行设计,能够灵活控制空域滤波器的带宽并抑制旁瓣电平。基于这种特性,设计的空域滤波器的主旁瓣比能够得到很大提高,从而提高测角性能。更重要的是,所提算法对发射和接收波束矩阵的结构进行设计,这种特殊的结构是构造实值信号模型的必要条件。最后,该文通过建立映射关系的方法对插值误差进行补偿。仿真结果验证了该算法的有效性。     

一种图像分割的快速不动点算法

该文在去除背景便能获得目标的分割思想之上,提出了一个凸的无约束最小化问题。证明了问题提出过程中添加惩罚项的合理性,并通过实验验证了证明结果。在最小化求解方面,应用次微分和近似算子的相关理论,构造了求解的不动点算子,进而结合Opial -averaged定理,给出了求解所提凸优化问题的不动点算法,并理论推导出了收敛条件,证明了算法的收敛性。与经典文献方法的对比实验表明所提方法分割结果更精确。同时实验显示该文算法比梯度下降法和分裂Bregman方法更快速。另外,所提算法对初始曲线和噪声有较好的鲁棒性。     

海上无人集群联合轨迹设计方法

针对海洋大数据收集场景,为提高数据收集效率,该文提出一种无人机和无人船联合数据收集方法。无人船在行驶过程中,通过放飞无人机并行驶到指定地点回收无人机,实现对目标海域内节点数据的高效收集。为最小化无人船和无人机工作时间,该文在无人机集群任务分配的基础上,引入连续悬停飞行(Successive-Hover-and-Fly, SHF)结构以实现低复杂度的联合轨迹优化。待优化问题受限于节点数据量和无人机速度,是难以求解的非凸问题。因此,该文提出了一个高效的连续凸近似技术迭代算法以得到次优解,并通过计算机仿真得以验证。     

Super-resolution mosaicing from MPEG compressed video

In this paper we address the problem of mosaic construction from MPEG 1/2 compressed video for the purpose of video browsing. State-of-the-art mosaicing methods work on raw video, but most video content is available in compressed form such as MPEG 1/2. Applying these methods to compressed video requires full decoding which is very costly. The resulting mosaic is in general too large to display on the screen and is thus inappropriate for the purpose of video browsing. Therefore, we directly extract very low-resolution frames from MPEG 1/2 compressed video for the mosaic construction and then apply a super-resolution (SR) method based on iterative backprojections in order to increase the mosaic resolution and its visual quality. Global motion to be used in the SR method for aligning and warping the frames is estimated from motion information contained in the compressed stream. We also use the estimated global motion in the blur estimation and in the choice of the degradation model used for the restoration in the SR algorithm. The method for the SR mosaic construction from MPEG 1/2 compressed video that we present in this paper is less costly than mosaic construction from full decoded video. Furthermore, the resulting mosaic size is more appropriate for the purpose of video browsing.     

基于JPEG序列的图像重建

该文提出了一种基于JPEG序列的图像重建方法。该方法在已有的单帧图像复原技术的基础之上,依据超分辨率重建的思想,将凸集投影(POCS)理论与迭代反投影(IBP)算法相结合,在频域内降低量化误差,修复离散余弦系数。此外,它还利用了最大后验概率(MAP)估计以及相应优化算法的特点,在去除高斯噪声的同时,保护边缘和细节信息。实验结果表明,该方法一方面能够抑制高比率压缩所造成的块效应和振铃效应,另一方面能较好地恢复图像的细节部分,有效地提高图像的清晰度。     

A fast temperature-aware fixed-outline floorplanning framework using convex optimization

With aggressive scaling of CMOS technology, it is essential to consider chip temperature in all design levels of digital systems to improve chip reliability and leakage power consumption. In this paper, we present a two phase fixed-outline floorplanning framework that attempts to reduce the peak-temperature of the chip. The first phase distributes evenly the available dead space between the floorplan blocks of a chip, so as to reduce the peak-temperature. The second phase employs a two-stage convex optimization formulation to perform fixed-outline floorplanning such that minimizes the peak-temperature while satisfying physical constraints. To mitigate the time and computational complexity of capturing the temperature behavior, we present a less computational expensive analogous formulation that approximates the temperature of a block by its corresponding power density. Although, the corresponding power density formulation exhibits lower complexity the experimental results demonstrate its high degree of accuracy. Moreover, this formulation manages to achieve significant improvements in terms of peak-temperature and runtime for almost all of the test cases. We investigate the trade-off between peak-temperature and area as well and provide conditions that result in a reasonable reduction of peak-temperature with minimum increase of the dead space.     

最佳中继协作通信系统的功率分配算法

 为提高基于最佳中继选择的协作通信系统的性能,提出了以最小化系统中断概率为目标的功率分配算法.首先建立了系统的优化模型并证明了待解的优化问题实质是凸优化问题,由此提出了最优功率分配算法并给出了算法步骤.其次,在此基础上提出了一种有效的次最优功率分配算法,该算法计算简单且仅需已知各个中继节点的平均信道状态信息,无需在传输中实时更新,因而不增加系统的额外开销.仿真结果表明,本文提出的最优算法和次最优算法所得到的功率分配方案与穷举搜索方法的结果非常接近;与等功率分配方案相比,这两种算法均能显著提高系统的中断概率性能.     

基于低秩张量补全的多声道音频信号恢复方法

多声道音频信号在采集、压缩、传输过程中可能造成音频数据丢失,为了确保给听众带来更真实的听觉感受,该文提出一种基于低秩张量补全的音频丢失数据恢复方法。首先,把多声道音频信号表示为一个张量;其次,把张量补全作为一个凸优化问题建模,利用松弛技术和变量分离技术得到闭合的增强拉格朗日函数;最后,通过交替迭代方法求解得到恢复的音频张量。在不同数据丢失率的实验中,通过与线性预测、加权优化的CANDECOMP /PARAFAC分解方法进行对比分析,表明利用张量补全方法具有更高的音频信号恢复精度,隐藏参考和基准的多激励测试结果也显示低秩张量补全方法能够有效地恢复多声道音频的丢失数据,从而获得更好的听觉效果。