A hardware design of a massive-parallel, modular NN-based vector quantizer for real-time video coding

This report describes the design of a modular, massive-parallel, neural-network (NN)-based vector quantizer for real-time video coding. The NN is a self-organizing map (SOM) that works only in the training phase for codebook generation, only at the recall phase for real-time image coding, or in both phases for adaptive applications. The neural net can be learned using batch or adaptive training and is controlled by an inside circuit, finite-state machine-based hard controller. The SOM is described in VHDL and implemented on electrically (FPGA) and mask (standard-cell) programmable devices.     

Kelvin Probe measurements of polymer coated gold substrates: Mechanism studies

The origin of work-function changes induced in polymer/metal structures (polyacrylic acid (PAA), ammonium and sodium polyacrylate) upon ammonia exposure is studied. To understand and explain the mechanism systematic measurements of uncoated and polymer coated metals are performed in Kelvin Probe (KP) setups. Additional data have been acquired through gravimetric and electrochemical experiments (cyclic voltammetry (CV)). The results obtained confirm that the field effect sensitive processes are taking place at the metal/polymer interface and allow their modelling. The intrinsic sensitivity of metallic samples upon ammonia exposure is influenced by the presence of coating polymers, by their ability to sorb water and to keep the metal surface wet or dry.     

一种实时倾斜跟踪处理机设计

针对自适应光学系统中的倾斜跟踪回路,设计了以TMS320C6701数字信号处理器和XC2V1500现场可编程门阵列为核心处理器、基于PCI 总线的新型实时倾斜跟踪处理机平台。阐述了该平台的基本组成,根据点目标介绍了质心跟踪算法的软件实现。实验结果表明,该平台能满足倾斜跟踪回路的实时性要求。     

液体流量计在线检测装置不确定度分析

液体流量计在线检测装置是由重庆市计量质量检测研究院研制出来的,主要用于现场检测关口水表、大口径流量计等用于贸易结算的流量计量器具。该装置克服了传统检测方法的不足、可操作性差的缺点,解决了关口水表不能现场进行检测的问题,具有实用性。该装置的建立满足了社会各企、事业单位的需要,并为政府计量管理和计量监督提供了科学依据。     

用FPGA实现基于PC／104的ARINC429接口模块

针对分离元件设计的总线接口电路复杂、可靠性差等缺点,采用FPGA技术结合高性能的HS-3282/HS-3182芯片组实现了基于PC/104的ARINC429总线接口模块。硬件上,该接口模块具有模块化强、电路简单、器件少、可靠性高等特点,软件上采用简洁方便的C语言实现多种工作方式下数据的收发功能。     

基于有向图模型的旅游领域命名实体识别

旅游领域命名实体识别是旅游知识图谱构建过程中的关键步骤,与通用领域的实体相比,旅游文本的实体具有长度长、一词多义、嵌套严重的特点,导致命名实体识别准确率低。提出一种融合词典信息的有向图神经网络（L-CGNN）模型,用于旅游领域中的命名实体识别。将预训练词向量通过卷积神经网络提取丰富的字特征,利用词典构造句子的有向图,以生成邻接矩阵并融合字词信息,通过将包含局部特征的词向量和邻接矩阵输入图神经网络（GNN）中,提取全局语义信息,并引入条件随机场（CRF）得到最优的标签序列。实验结果表明,相比Lattice LSTM、ID-CNN+CRF、CRF等模型,L-CGNN模型在旅游和简历数据集上具有较高的识别准确率,其F1值分别达到86.86%和95.02%。     

Ultrathin two-dimensional materials for photo- and electrocatalytic hydrogen evolution

Sustainable hydrogen production via photocatalytic, electrocatalytic, and synergetic photoelectrocatalytic processes has been regarded as an effective strategy to address both energy and environmental crises. Due to their unique structures and properties, emerging ultrathin two-dimensional (2D) materials can bring about promising opportunities to realize high-efficiency hydrogen evolution. This review presents a critical appraisal of advantages and advancements for ultrathin 2D materials in catalytic hydrogen evolution, with an emphasis on structure–activity relationship. Furthermore, strategies for tailoring the microstructure, electronic structure, and local atomic arrangement, so as to further boost the hydrogen evolution activity, are discussed. Finally, we also present the existing challenges and future research directions regarding this promising field.     

粮情测控技术及其发展

介绍了粮情测控系统及技术现状,简要论述了传感器技术、计算机技术、通信技术及控制技术的发展及在粮情测控系统中的应用情况,同时对粮情测控系统及技术的发展趋势进行了分析。相信随着科学技术的不断进步,此项技术不断发展,粮情测控技术必将向数字化、多功能继承化、自动调控、数据通信无线化和远程传输方向发展;而粮情测控产品的生产、安装和验收也必将趋于模块化、标准化和规范化。     

高速脉冲信号源的设计与实现

高速脉冲信号源是数据采集系统的重要组成部分,常用于系统的自检与测试中.根据实际要求,提出了高速运放调理和高速电子开关两种方案.分别对高速运放、信号幅值调理、驱动电路和高速电子开关的选择进行了研究,给出了具体的硬件电路设计.并结合相关理论,分析了信号在传输过程中出现的反射现象,提出了符合要求的解决方案.此高速脉冲信号源已成功应用于某采集设备的自检和测试中,效果良好.