基于运动传播和Isomap分析的三维人脸动画编辑与合成

提出一种三维人脸动画数据编辑与合成的有效方法,使用户可以在三维人脸模型上选定控制点,并在二维平面上指定表情动作的约束条件.根据人脸动画数据训练一个先验概率模型,将较少的用户约束传播到人脸网格的其他部分,从而生成完整生动的人脸表情;通过Isomap学习算法对三维人脸动画知识进行建模,并结合用户指定的关键帧拟合高维曲面上的平滑测地线,以自动合成新的人脸动画序列.实验结果表明,该方法可以直观地对人脸动画的生成进行交互式控制,并能生成较为逼真的表情动画.     

多径衰落信道环境中基于空时码构建的ST-OFDM系统

基于空时码发射分集技术,提出了一种ST-OFDM系统.对该系统的结构进行了简要的论述,并就空时码发射分集技术对系统性能的改善进行了分析.在此基础上,通过仿真证明了该系统的性能相对于传统的IEEE802.11a WLAN系统有了明显的提升,且新系统相对于传统系统在接收端不需要额外的付出,因此非常适合基于AP的移动无线网络环境.     

一种磁浮列车安全型计算机系统的设计与实现

本文以磁浮列车自动防护系统（ATP）为应用背景,对磁浮列车车载安全型计算机进行了研究。本文根据磁浮列车车载ATP系统对安全性的要求,对双机比较这一模式展开了研究,利用单片机作为实验平台,设计并实现了满足“故障-安全”原则的安全型计算机系统,并对各种可能出现的故障现象进行了实验验证。     

WEB2.0数据交换安全研究

伴随越来越多的网站步入Web2.0时代,在使用Ajax技术提供良好的用户体验的同时,一系列的安全问题也渐渐为人们所关注.本文就Web2.0应用中的数据交换环节.提出具体建议,以降低网络攻击的风险.     

A converter-assisting flexible switch topology and its control strategy北大核心

The rapid development of new energy in the distribution network increases its flexibility while making it more complicated to operate and manage the network. For this reason, a novel flexible switch, which is composed of integrated gate- commutated thyristors (IGCT) and parallel voltage converters, is proposed in this paper. Applying virtual synchronous generator (VSG) control on the parallel voltage converterensures smoothness and steadiness while switching from one feeder to another. Firstly, the proposed flexible switch is compared with the flexible multi-state switch and the mechanical switch on their main characteristics. Then, the operating features of the flexible switch before, during, and after feeder fault are analyzed, as well as its switching logic. Finally, the switching process between different feeders is simulated on a control in-the-loop experiment based on RT-Lab. The voltage and current waveforms are obtained under different operating modes and during flexible switching. The experiment resultsshow that the flexible switch has short-term voltage support capability and does not generate voltage and current impulses during the switching process. © 2023, Editorial Department of Electric Power Engineering Technology. All rights reserved.     

Ultraefficient Non-Doped Deep Blue Fluorescent OLED: Achieving a High EQE of 10.17% at 1000 cd m−2 with CIEy<0.08

The pursuit for efficient deep blue material is an ever-increasing issue in organic optoelectronics field. It is a long-standing challenge to achieve high external quantum efficiency (EQE) exceed 10% at brightness of 1000 cd m⁻² with a Commission International de L'Eclairage (CIE_y) <0.08 in non-doped organic light-emitting diodes (OLEDs). Herein, this study reports a deep blue luminogen, PPITPh, by bonding phenanthro[9,10-d]imidazole moiety with m-terphenyl group via benzene bridge. The non-doped OLED based on PPITPh exhibits an exceptionally high EQE of 11.83% with a CIE coordinate of (0.15, 0.07). The EQE still maintains 10.17% at the brightness of 1000 cd m⁻², and even at a brightness as high as 10000 cd m⁻², an EQE of 7.5% is still remained, representing the record-high result among non-doped deep-blue OLEDs at 1000 cd m⁻². The unprecedented device performance is attributed to the reversed intersystem crossing process through hot exciton mechanism. Besides, the maximum EQE of orange phosphorescent OLED with PPITPh as host is 32.02%, and remains 31.17% at the brightness of 1000 cd m⁻². Such minimal efficiency roll-off demonstrates that PPITPh is also an excellent phosphorescent host material. The result offers a new design strategy for the enrichment of high-efficiency deep blue luminogen.     

The Golden Touch by Light: A Finely Engineered Luminogen Empowering High Photoactivatable and Photodynamic Efficiency for Cancer Phototheranostics

Photoactivatable agent is a powerful tool in biomedicine studies due to high-precision spatiotemporal control of light. However, those previously reported agents generally suffer from short wavelength, fluorescence self-quenching effect, and the lack of photosensitizing property, which severely restrict their practical applications. To address these issues, molecular engineering of 1,4-dihydropyridine derivatives is conducted to obtain an optimized agent, namely TPA-DHPy-Py, which exhibits low oxidation potential, high photoactivation efficiency, and excellent type I/II combined photodynamic activity. Concurrently, its photoactivated counterpart is featured by aggregation-induced near-infrared emission and remarkable reactive oxygen species (ROS) production efficiency. Upon photoactivation, TPA-DHPy-Py is capable of precisely identifying cancer cells from co-culturing cancer cells and normal cells without the assistance of any extra targeting units, and in situ monitoring lipid droplets and endoplasmic reticulum alteration under ROS stress, as well as achieving fluorescent visualization of tumor in vivo with supremely high imaging contrast. Furthermore, the unprecedented performance on photodynamic cancer therapy is demonstrated by the significant inhibition of tumor growth. Therefore, the photoactivatable TPA-DHPy-Py with dual-organelle-targeted and excellent photodynamic activity associated with self-monitoring ability is highly promising for cancer theranostics in clinical trials.     

Freely Tailorable Yolk-Shell Encapsulation: Versatile Applications in Ultralow-k Dielectric,Drug Delivery Systems,and Catalysts

Herein, a facile, controllable, and versatile method is reported to prepare monodisperse yolk-shell and yolk-multishell silica nanoparticles (NPs) with mesoporous shells by a novel selective etching strategy. The mechanism of selective etching based on fluoride-silica chemistry is investigated in detail and thus provides a fundamentally novel principle for the fabrication of yolk-shell NPs. Specifically, this unprecedented and versatile synthesis strategy can be used to encapsulate essentially any silica-based, carbon-based, metal, metal oxide, or other possible NPs. Noteworthy is that most of the yolk-shell mesoporous silica (mSiO₂) NPs are prepared for the first time. To demonstrate the major structural and compositional advantages of the designed yolk-shell NPs, their applications in the fields of ultralow-dielectric constant (k) materials, drug delivery systems, and catalysts were explored. In detail, the lowest k value of the prepared yolk-shellordered mesoporous silica@mSiO₂/fluorinated polybenzoxazole composite films is 2.02; The obtained yolk-shell mSiO₂/C@mSiO₂/C NPs possess high hydrophilicity and pH-responsive sensitivity; The conversion of the catalytic reaction of the designed magnetic yolk-shell hollow Fe₃O₄@SiO₂/Au@mSiO₂ NPs at 20 min is 97% with a high conversion rate (92%) and recyclability even after 10 reuses. This innovative work lays a solid foundation for freely tailorable yolk-shell encapsulation and will greatly stimulate more efforts devoted to relevant research and development.     

基于动态门控特征融合的轻量深度补全算法

稠密深度图在自动驾驶和机器人等领域至关重要,但是现今的深度传感器只能产生稀疏的深度测量,所以有必要对其进行补全.在所有辅助模态中, RGB图像是常用且易得的信息.现今的许多方法都采用RGB和稀疏深度信息结合进行补全.然而它们绝大部分都是利用通道拼接或逐元素求和简单的对两种模态的信息进行融合,没有考虑到不用场景下不同模态特征的置信度.提出一种以输入深度稀疏分布为指导,结合双模态信息量的动态门控融合模块,通过动态产生融合权重的方式对两个模态特征进行更高效的结合.并且根据不同模态的数据特征设计了精简的网络结构.实验结果表明所提出模块和改进的有效性,提出的网络在两个有挑战性的公开数据集KITTI depth completion和NYU depth v2上,使用了很少的参数量达到了先进的结果,取得了性能和速度的优秀平衡.     

基于实时视频感知的虚拟体育交互系统

针对疫情常态化背景下,传统体育项目受场地、器材等限制,市场上相关产品价格昂贵、可扩展性不足等问题,提出了一种基于实时视频感知的虚拟体育交互系统.该系统设计视频数据采集模块和人体关节点提取模块,结合OpenPose获取人体的关节点坐标,实时捕捉人体手势以及肢体动作.动作语义理解模块包括运动动作理解和绘图动作理解.前者根据运动中肢体关节点的相对位置关系,识别运动动作语义.后者将手腕部关节点绘图动作轨迹生成为草图图像,使用AlexNet进行识别分类,解析为对应的绘制动作语义.该模型在边缘端设备的分类准确率为98.83%.采用基于Unity设计的草图游戏应用作为可视化交互界面,实现在虚拟场景中的运动交互.该系统使用实时视频感知交互方式实现居家运动健身,无需其他的外部设备,具有更强的参与度和趣味性.