一种新的自适应子波预滤波器设计

基于一般的紧支撑尺度函数,本文设计了一咎单参数且具有良好逼近性能的新的自适应ＦＩＲ预滤波器,分析和模拟结果与表明这种预波滤算法具有较高的逼近精度和较快的逼近速度。     

Video denoising via spatially adaptive coefficient shrinkage and threshold adjustment in Surfacelet transform domain

Many video denoising methods originated from methods designed for processing static two-dimensional images. Videos would be processed frame by frame, a process with a relatively high computational complexity, without taking into account the correlation information between frames. In this paper, a video denoising method using coefficient shrinkage and threshold adjustment based on Surfacelet transform (CSTA-ST) is proposed, which processes multiple frames of a video as an ensemble. Spatial correlation is used to define a weighted spatial energy. Each Surfacelet transform (ST) coefficient has a corresponding estimated energy value, in which the ST coefficients are grouped by. The similarity of the ST coefficients in a group determines the threshold of each ST coefficient. In addition, according to the neighborhood information of ST coefficients, the threshold is adjusted by a threshold adjustment factor. The coefficient shrinkage parameter is determined based on the adjusted threshold, and the ST coefficients are shrunk. Finally, the denoised video is obtained by the inverse ST using the shrunk coefficients. In experiments, video sequences with noise are tested, and the denoised results of the proposed method are compared with that of current denoising methods. The experimental results show that the proposed method significantly improves the peak signal-to- noise ratio (PSNR) and the structural similarity (SSIM) for various levels of noise and motion, and the ideal denoised visual effect is obtained.     

用纳米棒ZnO作模板生长无支撑的AlN纳米晶

用金属有机物气相外延在纳米棒ZnO模板上沉积AlN薄膜.SEM测试表明该薄膜形成了一种倾倒纳米棒的表面.而GIXRD测试进一步证实它是纤锌矿结构的AlN,晶粒尺度约为12nm,接近于ZnO纳米棒的直径(30nm).这意味着纳米棒结构的ZnO能限制AlN的横向生长.此外,高温下用H2刻蚀ZnO直接在生长中实现了外延层的剥离.最终得到了无支撑的AlN纳米晶,完整无破损的区域约为1cm×1cm.定义这个生长机制为"生长-刻蚀-合并"过程.     

2.4GHz无线技术标准及ZigBee抗干扰性能

介绍了ZigBee和其他几种工作于2.4GHz ISM频段的短距离无线通信技术标准,Wi-Fi、蓝牙和无线USB,对ZigBee本身的抗干扰性能以及与其他无线技术的共存进行了分析,讨论了如何保证ZigBee避免干扰和改善其共存性能.     

基于稀疏编码和集成学习的多示例多标记图像分类方法

该文基于稀疏编码和集成学习提出了一种新的多示例多标记图像分类方法。首先,利用训练包中所有示例学习一个字典,根据该字典计算示例的稀疏编码系数;然后基于每个包中所有示例的稀疏编码系数计算包特征向量,从而将多示例多标记问题转化为多标记问题;最后利用多标记分类算法进行求解。为了提高分类器的泛化能力,对多个分类器进行集成。在多示例多标记图像数据集上的实验结果表明所提方法与其它方法相比有更好的性能。     

Multichannel Sonocatalysis Amplifiers Target IDH1-Mutated Tumor Plasticity and Attenuate Ros Tolerance to Repress Malignant Cholangiocarcinoma

Tumor adaptation-originated tumor tolerance that compensatory mechanisms (e.g., isocitrate dehydrogenase (IDH) mutation) jointly shape is the dominant obstacle of ROS therapy. Currently, targeting a single pathway fails to fundamentally reverse the complex milieu and diminish tumor adaptation. Herein, a multichannel sonocatalysis amplifier is engineered via one-pot gas diffusion method to attenuate IDH1-mutated cholangiocarcinoma plasticity and tolerance to ROS therapy, wherein triptolide and IR780 are co-loaded in DSPE-mPEG-modified CaCO₃ nanoparticles. Triptolide can blockade Nrf2 to cut off glutathione biosynthesis via blockading proteomic communication, and disrupt redox homeostasis to potentiate IR780-mediated sonocatalytic ROS production. ROS-induced mitochondria damages disrupt Ca²⁺ homeostasis and in turn aggravate ROS accumulation, which cooperates with sonocatalysis and Nrf2 blockade to reprogram mitochondrial energy and substance metabolism (e.g., adenosine triphosphatase and glutathione), hinder DNA self-repair, and impair IDH1-mutation-asired tolerance. Systematic experiments support that these actions in such multichannel sonocatalysis amplifiers indeed disrupt Ca²⁺/redox homeostasis to disarm robust tumor plasticity and IDH1-mutation-induced tolerance to sonocatalysis therapy against IDH1-mutated cholangiocarcinoma progression. Briefly, the sonocatalysis amplifiers pave a comprehensive avenue to reprogram tumor metabolism, target tumor vulnerability, and attenuate tumor plasticity against genomic instability-raised treatment adaptation.     

Tunable Optical Mode Ferromagnetic Resonance in FeCoB/Ru/FeCoB Synthetic Antiferromagnetic Trilayers under Uniaxial Magnetic Anisotropy

Ferromagnetic resonance (FMR) is one of the most important characteristics of soft magnetic materials, which practically sets the maximum operation speed of these materials. There are two FMR modes in exchange coupled ferromagnet/nonmagnet/ferromagnet sandwich films. The acoustic mode has relatively lower frequency and is widely used in radio‐frequency/microwave devices, while the optical mode is largely neglected due to its tiny permeability even though it supports much higher frequency. Here, a realistic method is reported to enhance the permeability in the optical mode to an applicable level. FeCoB/Ru/FeCoB trilayers are carefully engineered with both uniaxial magnetic anisotropy and antiferromagnetic interlayer exchange coupling. This special magnetic structure exhibits a high optical mode frequency up to 11.28 GHz and a maximum permeability of 200 at resonance. An abnormally low inverse switch field (<200 Oe, less than 1/5 of the single layer) is observed which can effectively switch the system from optical mode with higher frequency into acoustic mode with lower frequency. The optical mode frequency and inverse switch field can be controlled by tailoring the interlayer coupling strengths and the uniaxial anisotropy fields, respectively. The tunable optical mode resonance thus can increase operation frequency while reduce operation field overhead in FMR based devices.     

Despeckling SAR images based on a new probabilistic model in nonsubsampled contourlet transform domain

A method for synthetic aperture radar (SAR) image despeckling based on a probabilistic generative model in nonsubsampled contourlet transform (NSCT) domain was proposed. The shrinkage estimator in NSCT domain consists of a new type of likelihood ratio and prior ratio, both of which are dependent on the estimated masks for the NSCT coefficients. While the previous probabilistic approaches are restricted to parametric models, the limitation is eliminated and the hybrid density model is applied in this paper. The suggested approach does not make heavy assumptions on the NSCT coefficient distribution, so that it can handle complex NSCT coefficient structures. The likelihood ratio is composed of the hybrid density, and the prior ratio is equipped with the selective neighborhood systems to enhance the detail information. The method can effectively adapt the shrinkage estimator to the redundancy property of the NSCT. The proposed approach was applied to real SAR images despeckling and compared through the SAR image vision effect, the equivalent number of looks, and the edge sustain index. Experimental results show that the proposed approach outperforms previous works involved in the paper with the better despeckling result and edge preservation.     

基于像素相似性与Graph-Cut的图像自动分割

采用Graph cut的图像分割分为交互式分割与自动分割,为了弥补交互式分割需要用户的参与、分割结果依赖种子点的选取与模型的建立、需要用户的进一步修正才能得到满意结果等不足,介绍基于相似性的自动图像分割。该方法采用Mean-shift平滑图像、转换成YCbCr空间后将图像分割成若干区域块等方法,建立合适的能量函数后利用最大流/最小割定理求最小割。并通过实验证明,该方法在自动性、准确性、时间效率方面都获得较好结果。     

基于变异的紧凑遗传算法在农药检测中的应用

为了简化农药检测的预测模型,提高模型预测精度,采用红外光谱技术结合基于变异的紧凑遗传算法对波长变量进行筛选,一定程度上减少了无信息变量和干扰变量。通过不同算法选择的波长变量建立预测模型,mCGA得到的预测均方根偏差平均值是0.198,而与mCGA比较的紧凑遗传算法、简单遗传算法得到的预测均方根偏差平均值分别为0.241、0.289,mCGA具有最小误差。结果表明,采用mCGA进行变量选择,能有效提高模型收敛速度及模型准确度,实现农药含量快速高效的检测。