Grain‐Boundary “Patches” by In Situ Conversion to Enhance Perovskite Solar Cells Stability

The power conversion efficiency of organic–inorganic hybrid perovskite solar cells has increased rapidly, but the device stability remains a big challenge. Previous studies show the grain boundary (GB) can facilitate ion migration and initiate device degradation. Herein, methimazole (MMI) is employed for the first time to construct a surface “patch” by in situ converting residual PbI₂ at GBs. The resultant MMI–PbI₂ complex can effectively suppress ion migration and inhibit diffusion of the metal electrodes. The origin of the surface “patch” effect and their working mechanisms are investigated experimentally and theoretically at the microscopic level. It hence demonstrates a simple and effective method to prolong the device stability in the context of GB engineering, which could be extensively applied to perovskite‐based optoelectronics.     

基于BJT特性的数字温度传感器发展现状及趋势

随着工业物联网趋向数字化、智能化和集成化发展,控制系统需要感知的物理量规模和复杂度都迅速提升。其中数字温度传感器能直接将温度信息转换为数字信号,具有低成本、低功耗、面积小、数字输出等多种优点,可以实时监测系统温度数据,并与反馈机制协同进行反馈调节,目前已经得到广泛应用。在各类数字温度传感器中,基于CMOS工艺寄生三极管(BJT)感温的数字温度传感器在制造工艺上更容易实现,且具有高稳定性和高精度,是工业界产品首选方案。聚焦基于BJT特性实现感温的数字温度传感器,从学术研究成果、工业产品两方面总结其技术路线、发展现状和趋势,为后续温度传感器研究提供参考。     

消防安全管理现状与消防监督管理模式探讨

文章从我国消防安全管理的现状出发,简要分析了现存的消防安全管理问题,并就消防监督管理工作的创新模式提出意见,以期为相关工作者带来一定的参考价值。     

英国约克郡数字电视发展见闻与思考

介绍了英国约克郡数字电视发展见闻、目前的技术研究重点和新数字业务的开拓情况,注意到其促进数字产业发展的组织形式,即政府推动产学研结合加速技术研发和产品推广应用的先进方法.     

跨层协同注意和通道分组注意的细粒度图像分类

细粒度图像分类的主要挑战在于类间的高度相似性和类内的差异性. 现有的研究多数基于深层的特征而忽略了浅层细节信息, 然而深层的语义特征由于多次卷积和池化操作往往会丢失大量的细节信息. 为了更好地整合浅层和深层的信息, 提出了基于跨层协同注意和通道分组注意的细粒度图像分类方法. 首先, 通过ResNet50加载预训练模型作为骨干网络提取特征, 由最后3个阶段提取的特征以3个分支的形式输出, 每一个分支的特征通过跨层的方式与其余两个分支的特征计算协同注意并交互融合, 其中最后一个阶段的特征经过通道分组注意模块以增强语义特征的学习能力. 模型训练可以高效地以端到端的方式在没有边界框和注释的情况下进行训练, 实验结果表明, 该算法在3个常用细粒度图像数据集CUB-200-2011、Stanford Cars和FGVC-Aircraft上的准确率分别达到了89.5%、94.8%和94.7%.     

Ultra-thin Body Buried In0.35Ga0.65As Channel MOSFETs with Extremely Low Off-current on Si Substrates

In this paper, we investigated the electrical properties of the Metal-oxide-semiconductor gate stack of Ti/Al2O3/InP under different annealing conditions. A min...     

Studies on Lotus Genomics and the Contribution to Its Breeding

Lotus (Nelumbo nucifera), under the Nelumbonaceae family, is one of the relict plants possessing important scientific research and economic values. Because of this, much attention has been paid to this species on both its biology and breeding among the scientific community. In the last decade, the genome of lotus has been sequenced, and several high-quality genome assemblies are available, which have significantly facilitated functional genomics studies in lotus. Meanwhile, re-sequencing of the natural and genetic populations along with different levels of omics studies have not only helped to classify the germplasm resources but also to identify the domestication of selected regions and genes controlling different horticultural traits. This review summarizes the latest progress of all these studies on lotus and discusses their potential application in lotus breeding.     

基于自适应阈值分割改进Canny算子的管道边缘检测方法研究

针对工厂中管道破损位置无法通过机器视觉准确判断的问题,提出一种基于自适应阈值分割改进Canny算子的管道边缘检测方法。该方法从滤波方式、梯度方向以及阈值分割角度对采集图像进行处理,首先采用采样-自适应中值滤波+双边滤波代替传统Canny算子中的高斯滤波,减少图像边缘信息丢失并去除图像中的噪声,然后增加梯度幅值的计算来更好地检测不同方向的边缘信息,最后为避免人工选取阈值效果不佳的情况,采用最大类间方差 (OTSU)阈值分割算 法进行阈值的自适应选取。实验表明,该方法相比于传统Canny算子的图像信噪比提升28.22%,边缘点数提升39.97%,四连通道数提升11.52%,八连通道数提升5.92%,提取特征完整且连续性较好,实现了对管道图像中破损情况的有效检测。     

透镜球差对拉盖尔-高斯涡旋光束聚焦特性的影响及改善方法

涡旋光场调控对涡旋光在激光通信、生物操控等领域的应用研究具有重要意义。为了揭示透镜球差对拉盖尔-高斯(Laguerre-Gaussian,LG)涡旋光束传输特性的影响,本文基于菲涅尔衍射理论,模拟研究了不同透镜球差系数下的LG光束聚焦场的强度、相位以及螺旋谱分布。结果表明:光学系统初级球差系数的增大恶化了LG光束聚焦场的强度分布和螺旋相位,弥散了轨道角动量(orbital angular momentum,OAM) 。由此,本文提出了一种光阑补偿方案,通过在透镜前添加大小合适的圆孔光阑,以消除透镜像差。光阑补偿后的LG光束聚焦场由星芒状空心分布恢复为空心圆环分布,OAM态保持单一稳定。本研究工作为改善透镜像差对涡旋光的传输影响提供了解决方案。     

Ordinal Regression for Information Retrieval

This letter presents a new discriminative model for Information Retrieval （IR）, referred to as Ordinal Regression Model （ORM）. ORM is different from most existing models in that it views IR as ordinal regression problem （i.e. ranking problem） instead of binary classification. It is noted that the task of IR is to rank documents according to the user information needed, so IR can be viewed as ordinal regression problem. Two parameter learning algorithms for ORM are presented. One is a perceptron-based algorithm. The other is the ranking Support Vector Machine （SVM）. The effectiveness of the proposed approach has been evaluated on the task of ad hoc retrieval using three English Text REtrieval Conference （TREC） sets and two Chinese TREC sets. Results show that ORM significantly outperforms the state-of-the-art language model approaches and OKAPI system in all test sets; and it is more appropriate to view IR as ordinal regression other than binary classification.