NOON State Generation with Phonons in Acoustic Wave Resonators Assisted by a Nitrogen‐Vacancy‐Center Ensemble

Since the quality factor of an acoustic wave resonator (AWR) reached 10¹¹, AWRs have been regarded as a good carrier of quantum information. In this paper, a scheme to construct a NOON state with two AWRs assisted by a nitrogen‐vacancy‐center ensemble (NVE) is proposed. The two AWRs cross each other vertically, and the NVE is located at the center of the crossing. By considering the decoherence of the system and using resonant interactions between the AWRs and the NVE, and the single‐qubit operation of the NVE, a NOON state can be achieved with a fidelity higher than 98.8% when the number of phonons in the AWR is $N\le 3$.     

高光谱图像识别霉变花生的光谱特征分析与指数模型构建

霉变花生极有可能含强致癌物质-黄曲霉素，快速识别并分离霉变花生可从源头上阻止其进入食物链，并降低人类摄入黄曲霉素的风险。利用可见光-近红外高光谱数据，通过光谱分析确定能有效识别霉变花生的光谱特征或指数模型。共获取霉变花生样本253个，健康花生247个，并取其霉变(或健康)部位的均值光谱。在对光谱进行连续统去除后，首先对其求取了不同步长的一阶微分，并在可分性较优的光谱区域计算了Area_500～650指数；其次，用连续小波变换提取了光谱的形状和位置信息，并利用Index_cwt指数识别霉变花生样本。结果显示，指数Area_500～650的J-M距离为1.95，Index_cwt模型的J-M距离为1.99，表明霉变和健康花生在构建的指数模型Area_500～650和Index_cwt的特征空间可分性均较优。     

Digital Interferometric Measurement of Forced Convection Heat Transfer in a Miniature Rectangular Channel

Investigations have been performed on convective heat transfer in water flowing through mini-channels using the non intrusive technique of laser interferometry coupled with digital image processing. Optical glass channels, fabricated with metallic heating surfaces, were studied using a Mach-Zehnder interferometer configuration. Fringe patterns captured using a high-sensitivity CCD camera were analyzed digitally based on a calculation method developed for the liquid medium. Results of parametric studies were compared and contrasted with relevant theoretical solutions from the literature. Indication of the onset of turbulence at Reynolds numbers smaller than the conventional transition Reynolds number for large channels has also been noticed in the experimental investigation.     

Die Korrektur des Bildes der Anger-Kamera (Teil II)

Ausgehend von der im ersten Teil begründeten Notwendigkeit, die mit der Szintillationskamera Impulsverteilung korrigieren zu müssen, wird in der vorliegenden Arbeit der Programmablanfplan für diese Korrekturrechnung mit Hilfe der elektronischen Datenverarbeitung (R 300) erläutert. Im gegenwärtigen Stadium der Entwicklung eines umfangreichen Konzepts zum Einsatz der EDV für die Szintigraphie – an der alle Nuklearmedizinischen Abteilungen der DDR beteiligt sind – erfolgt bei den Autoren die Ein- und ausgabe der Daten über Lochstreifen. Die Wirkung der Korrekturrechnung wird am homogenen Phantom selbst und an einer Pankreasunfersuchung demonstriert.     

Fingerprint image enhancement using data driven Directional Filter Bank

A new method based upon data driven tool, principal component analysis (PCA), for fingerprint enhancement is proposed in this paper. PCA is a very useful statistical technique that has found application in many different fields like image compression, face recognition and is commonly used for finding patterns in data of high dimension. In the proposed method, the input image is first decomposed into directional images using decimation free Directional Filter Bank (DDFB). Then these directional images are normalized. A data driven technique PCA is applied to these normalized directional fingerprint images, which gives the PCA filtered images. These are basically directional images. Then these directional images are reconstructed into one image which is the enhanced one. Simulation results are included illustrating the capability of the proposed method.     

声学超表面的非对称声分束特性研究

非对称声分束超表面是由人工微单元结构按照特定序列构建的二维平面结构,可将垂直入射的声波分成两束传播方向和分束比自由调控的透射波,在声功能器件设计及声通信领域具有广泛的应用前景。本文系统研究了一种实现非对称声分束的设计理论和实现方法,基于局域声功率守恒条件研究了声分束器的设计理论、阻抗矩阵分布、法向声强分布、声压场分布等。利用遗传算法对四串联共振腔结构进行参数优化实现了声分束器所需的阻抗矩阵分布,声压场分布表明声波入射到声分束器后在入射侧激发出两列传播方向相反且幅值和衰减系数均相同的表面波,实现了入射侧与透射侧的局域声功率相互匹配。声波经过声分束器后被分为两束透射波,两束透射波的折射角和透射系数与理论值十分吻合,证明了设计理论及实现方法的正确性和可行性。本文的研究工作可以为新型非对称声分束结构设计提供理论参考、设计方法和技术支持,并促进其在工程领域的实际应用。     

医学影像处理的深度学习可解释性研究进展

随着医学影像数据的迅速增长，传统的影像分析方法给医生带来巨大挑战。利用计算机视觉技术提供自动或半自动辅助诊断，可大大缓解人工阅片压力，提高诊断的准确性，促进医疗流程的标准化建设等。目前，深度学习卷积神经网络在医学影像处理中已取得不俗表现，但深度学习“黑匣子”的不可解释性阻碍了智能医疗诊断的发展。为增强对医学影像数据处理的深度学习可解释性的了解，对近几年相关研究进展进行了综述。首先，综述了深度学习在医学领域的应用现状及面临的问题，对神经网络的可解释性内涵进行了讨论；然后，从现有深度学习可解释性的常见方法出发，重点讨论了医学影像处理的深度学习可解释性研究进展；最后，探讨了医学影像处理的深度学习可解释性的发展趋势。     

基于快速神经网络架构搜索的鲁棒图像水印网络算法

为解决深度学习在图像水印算法中计算量大且模型冗余的问题，提高图像水印算法在抵抗噪声、旋转和剪裁等攻击时的鲁棒性，提出基于快速神经网络架构搜索（neural architecture search，NAS）的鲁棒图像水印网络算法。通过多项式分布学习快速神经网络架构搜索算法，在预设的搜索空间中搜索最优网络结构，进行图像水印的高效嵌入与鲁棒提取。首先，将子网络中线性连接的全卷积层设置为独立的神经单元结构，并参数化表示结构单元内节点的连接，预先设定结构单元内每个神经元操作的搜索空间；其次，在完成一个批次的数据集训练后，依据神经元操作中的被采样次数和平均损失函数值动态更新概率；最后，重新训练搜索完成的网络。水印网络模型的参数量较原始网络模型缩减了92%以上，大大缩短了模型训练时间。由于搜索得到的网络结构更为紧凑，本文算法具有较高的时间性能和较好的实验效果，在隐藏图像时，对空域信息的依赖比原始网络更少。对改进前后的2个网络进行了大量鲁棒性实验，对比发现，本文算法在CIFAR-10数据集上对抵抗椒盐噪声和旋转、移除像素行（列）等攻击优势显著；在ImageNet数据集上对抵抗椒盐高斯噪声、旋转、中值滤波、高斯滤波、JPEG压缩、裁剪等攻击优势显著，特别是对随机移除行（列）和椒盐噪声有较强的鲁棒性。     

Time reversal acoustic communication receivers: DSP implementation and fast channel estimation

A communication system is implemented on digital signal processors (DSPs) for the underwater acoustic environment. The implemented receiver uses time reversal multi-channel combining followed by a single-channel decision feedback equalizer. Periodic channel estimation is employed to track the channel fluctuations. These techniques are used to mitigate time-varying inter-symbol interference, which is the main challenge in the underwater acoustic channel at operating frequencies greater than 10 kHz. Various optimization tasks are performed to reduce the receiver computational complexity. A fast implementation of the matching pursuit algorithm is tested on the DSP platform. Its performance, in terms of accuracy and run-time, is compared with that of the basic matching pursuit algorithm. Experimental results of the transmission and demodulation of binary phase-shift keying signals at three different symbol rates were obtained in the local Delaware Bay. The low bit error rates demonstrate the effectiveness of our implementation.