数据融合技术在无线传感器网络中的应用

在大规模的无线传感器网络中,传输数据量巨大,必然存在着数据传输可靠性、拥塞以及能耗等问题,高效的数据融合技术能够有效的解决这些问题。本文结合分簇路由算法的特征,采用两层融合技术,首先簇内节点与簇首节点的融合,簇内节点根据阈值来判断是否需要发送数据,簇首节点根据接收到的数据,进行数据一致性检验,剔除异常数据,第二层采用BP神经网络算法对簇首节点与基站的融合,得到所需要的结果。实验表明,进行融合后的数据可靠性高,较大减少了数据的传输量与冗余度、降低了能量的消耗,从而提高了整个网络的性能。     

基于深度学习的齿轮视觉微小缺陷检测

针对齿轮视觉微小缺陷,采用一种基于深度学习算法的Mask R-CNN网络进行检测,并对网络进行相应地优化调整.首先,通过比较5种残差神经网络检测效果,选择resnet-101作为图像共享特征提取网络.然后,剔除特征金子塔网络中对特征图P₅进行的不合理的3×3卷积,缺齿检出率指标相应得到提升.最后,为了对候选区域网络进行有效的训练,根据设计的样本标注方案中小范围波动的标注尺寸,设置合适的anchors大小以及宽高比.最终,经过优化的Mask R-CNN网络达到了98.2%缺齿检出率.     

Label‐free identification of the microstructure of rat spinal cords based on nonlinear optical microscopy

The spinal cord is a vital link between the brain and the body and mainly comprises neurons, glial cells and nerve fibres. In this work, nonlinear optical (NLO) microscopy based on intrinsic tissue properties was employed to label‐freely analyze the cells and matrix in spinal cords at a molecular level. The high‐resolution and high‐contrast NLO images of unstained spinal cords demonstrate that NLO microscopy has the ability to show the microstructure of white and grey matter including ventral horn, intermediate area, dorsal horns, ventral column, lateral column and dorsal column. Neurons with various sizes were identified in grey matter by dark spots of nonfluorescent nuclei encircled by cytoplasm‐emitting two‐photon excited fluorescence signals. Nerve fibres and neuroglias were observed in white matter. Besides, the spinal arteries were clearly presented by NLO microscopy. Using spectral and morphological information, this technique was proved to be an effective tool for label‐freely imaging spinal cord tissues, based on endogenous signals in biological tissue. With future development, we foresee promising applications of the NLO technique for in vivo, real‐time assessment of spinal cord diseases or injures.     

乳腺肿瘤红外热成像及其与超声影像的比较

采用红外热成像系统联合彩色多普勒超声仪对8例女性乳腺肿瘤患者进行扫描成像,获取乳腺肿瘤温度分布等红外热辐射特征及其乳腺肿瘤的大小、内部血流等超声影像特征,探讨乳腺肿瘤的红外热图像特征,同时将红外热图像分析结果与超声诊断结果进行比较,结果表明两种影像技术的吻合度较好,研究结果具有重要的应用价值。     

Near-field and far-field characteristics of a MQW waveguide

Definitions of beam half-width and divergence half-angle according to different kinds of methods were investigated. Results of these definitions are given for the fundamental mode and numerical calculation proves that some of them are more appropriate for describing the propagation characteristics of a multiple quantum well (MQW) planar waveguide.     

Quantitative biomarkers of human skin photoaging based on intrinsic second harmonic generation signal

Collagen change is a major feature in the photoaged human skin. Here, we present the use of intrinsic second harmonic generation (SHG) signal as a novel means to quantify collagen change with photoaging. We obtain the SHG images of the superficial dermis from ex vivo the cheek skin and the abdomen skin of eight patients aged 55–60 years. The results show that SHG signal can quantitatively reveal collagen change between normal and photoaged human skin in three dimensions. By comparing normal with photoaged dermis, there are significant differences in the collagen content and fine structure, providing substantial potential to be applied in vivo for the clinical diagnosis of human skin photoaging. SCANNING 35: 273‐276, 2013. © 2012 Wiley Periodicals, Inc.     

High Throughput Blood Analysis Based on Deep Learning Algorithm and Self-Positioning Super-Hydrophobic SERS Platform for Non-Invasive Multi-Disease Screening

Blood analysis is crucial for early cancer screening and improving patient survival rates. However, developing an effective strategy for early cancer detection using high-throughput blood analysis is still challenging. Herein, a novel automatic super-hydrophobic platform is developed together with a deep learning (DL)-based label-free serum and surface-enhanced Raman scattering (SERS), along with an automatic high-throughput Raman spectrometer to build an effective point-of-care diagnosis system. A total of 695 high-quality serum SERS spectra are obtained from 203 healthy volunteers, 77 leukemia M5, 94 hepatitis B virus, and 321 breast cancer patients. Serum SERS signals from the normal (n = 183) and patient (n = 443) groups are used to assess the DL model, which classify them with a maximum accuracy of 100%. Furthermore, when SERS is combined with DL, it exhibits excellent diagnostic accuracy (98.6%) for the external held-out test set, indicating that this method can be used to develop a high throughput, rapid, and label-free tool for screening diseases.     

Two-photon fluorescence and second-harmonic generation imaging of collagen in human tissue based on multiphoton microscopy

Multiphoton microscopic imaging of collagen plays an important role in noninvasive diagnoses of human tissue. In this study, two-photon fluorescence and second-harmonic generation (SHG) imaging of collagen in human skin dermis and submucosa of colon and stomach tissues were investigated based on multiphoton microscopy (MPM). Our results show that multiphoton microscopic image of collagen bundles exhibits apparently different pattern in human tissues. The collagen bundles can simultaneously reveal its SHG and two-photon excited fluorescence images in the submucosa of colon and stomach, whereas it solely emit SHG signal in skin dermis. The intensity spectral information from tissues further demonstrated the above results. This indicates that collagen bundles have completely different space arrangement in these tissues. Our experimental results bring more detailed information of collagen for the application of MPM in human noninvasive imaging.     

双驱动端口的中高频RFID双层天线设计

为了提升天线线圈在空间中产生的磁场强度,提出一种双驱动端口的双层天线方案.首先,从理论出发介绍了设计双层天线时应该考虑的各个参数;其次,改变双层天线的参数,通过三维电磁仿真软件Ansoft HFSS进行建模仿真,探究两线圈的匝数、间距、半径以及相对位置对双层天线的影响;最后,对仿真结果进行分析,找出了最优的双层天线设计方案,对实际设计的中高频RFID读写器天线具有一定的参考价值.