浅谈模式识别在流量建模中的应用

本文简介了模式识别的概念、研究场景分类以及常用的模式识别方法,并针对移动通信系统性能仿真中的信源流量建模环节,探讨了模式识别在传感器网络流量建模中的应用。     

三元气敏阵列和有机溶剂识别

本文用ＳｎＯ２气敏元件构造了三元阵列，使用ｋ－最近邻和加权ｋ－最近邻两种分类器，结果表明，前一分类器对乙醇和汽油的识别率分别为４０％和９０％，后一类分类器则分别为７０％和９０％，模式识别是改善气敏元件选择性的有效途径。     

气敏阵列及其应用

介绍了气敏阵列器件的工作原理、硬件实现方案和模式识别方法，重点描述了气敏阵列产品在各个行业中的应用。同时指出了在气敏阵列应用中存在的主要问题。     

传感探测技术的发展动态

传感探测技术通过声、光、电、热等物理量的变化,实现对物体的探测、识别、跟踪和定位。它是一门涉及传感技术、微机械和微电子技术、信号处理技术、电路与系统、模式识别、神经网络等多种学科的综合性技术。传感探测系统通常由各种传感器、高速实时信号处理器、数据通信电路、模式识别和管理软件等组成。随     

基于模糊模式识别的车型识别系统

车型的识别在智能交通系统中有重要作用,文中主要介绍了一种基于压力传感器和红外传感器来提取车辆特征,基于模糊模式识别方法进行识别的车型识别系统.     

具有领域独立性的通用模式识别方法的研究

统计方法是当前广泛采用的模式识别方法，但是这种方法的专业性较强，缺乏通用性。针对这种现状文章提出了基于结构模式识别理论的一种通用模式识别方法，该方法使用一系列结构检测子来提取信号特征．具有领域独立性。设计了对比实验，通过同传统的统计模式识别方法的比较验证了通用模式识别方法的有效性．通过两个不同领域的模式识别问题验证了该方法的领域独立性和有效性。     

混合型磁光电流传感器灵敏度的提高

选择高Ｖｅｒｄｅｔ常数的磁光材料，增加光束环绕电流的环咱数是提高磅光电流传感器灵敏度的主要方法。本文介绍了利用频率分离的比较测量法来消除温度对高Ｖｅｒｄｅｔ常数铁磁性材料的影响。设计了双环路及多环路块状传感头结构，并对双环咱结构传感头进行了实验测试。     

基于光纤光栅传感系统的入侵信号识别

介绍了一种基于光纤光栅传感器的周界检测系统中的信号处理方法,即时频数据分析法。在时域信号分析的基础上,通过将时域信号进行频域转换来进行入侵信号的模式识别及分析,并利用MATLAB仿真计算结果加以验证。试验结果表明该方法可以提升系统识别不同的入侵行为的能力,解决了纯时域分析下难实现的模式识别问题。     

基于模糊控制的电磁智能车设计

智能车控制是涵盖自动控制、模式识别、传感技术、计算机、机械与汽车等多个学科的复杂系统.为了实现对其很好的控制,文章提出了基于模糊控制的电磁引导的智能车控制方案.实际结果表明智能车运行性能优良.     

一种利用SPATE效应近红外变量的新型光纤应力／应变传感器

本文介绍一种用于监测热材料的新型光纤应力／应变传感器，光纤能收集来自于被加热的材料样品上的灰体辐射，当样品经受瞬时应力时，由于在样品上温度的绝热变化，辐射被调制，用一只近红外ＧａＩｎＡｓ光测器，经过石英光纤可以监测这种调制，这也是ＳＰＡＴＥ方法在近红外区的首次验证。     

基于SAW传感器阵列的混合气体定性分析

针对野外文物古迹环境的多变性、传感器气敏性失效以及腐蚀性气体定性不精确等实际情况,结合物联网技术和模式识别技术,在综合考虑准确度和项目实际要求的情况下,设计了一种基于声表面波(SAW)传感器阵列的模式识别算法,并通过MATLAB对其进行仿真和验证。结果证明,将传感器阵列输出的数据输入到有6个隐层的神经单元的单隐层BP神经网络中进行训练,预测效果最好,对腐蚀性气体的识别率达到了95%左右,提高了野外微气象环境下腐蚀性气体的监测水平。     

The fifth sense-on the scent of the electronic nose

The author describes how electronic sensors for scent and odours work. Gases produce changes in the properties of an array of sensors, resulting in electrical signals, which are then amplified, linearised for gas concentration and digitised before being fed into a pattern-recognition system. The gas sensitive layer is tin oxide with catalytic additives for detection of particular gases. The author discusses how different sensors combined with pattern recognition techniques, can discriminate between different odours. The different types of sensor and their application to gas testing systems are also discussed     

A high precise E-nose for daily indoor air quality monitoring in living environment

E-nose, whose major components include a sensor array and a pattern recognition algorithm, is considered to be a potential way to balance the trade-off between cost and accuracy for daily indoor air quality monitoring in living environment. In this paper, we presented a high precise E-nose for such application. QS-01 from FIS, TGS2600 and TGS2602 from FIGARO, temperature and humidity sensor SHT10 are selected to compose the sensor array. Back Propagation (BP) nueral network, the typical machine learning algorithm is used to be the pattern recognition algorithm of the E-nose. The performance comparison between the proposed E-nose and other E-nose solutions shows the improvement.     

化学战剂检测器的研究

研究了一种基于声表面波传感器阵列、后向人工神经网络(BP-ANN)模式识别软件和信号处理硬件的声表面波化学战剂检测器.传感器阵列输出的信号首先在FPGA中进行计数,然后通过MCU进行BP-ANN模式识别分析.根据分析的结果,MCU控制检测器发出报警信号.     

Colorimetric Sensor Arrays System Based on FPGA for Image Recognition

A FPGA-based image recognition system is designed for colorimetric sensor array in order to recognize a wide range of volatile organic compounds. The gas molecule is detected by the responsive sensor array and the responsive image is obtained. The image is decomposed to RGB color components using CMOS image sensor. An embedded image recognition architecture based on Xilinx Spartan-3 FPGA is designed to implement the algorithms of image recognition. The algorithm of color coherence vector is discussed in detail compared with the algorithm of color histograms, and experimental results demonstrate that both of the two algorithms could be analyzed effectively to represent different volatile organic compounds according to their different responsive images in this system.     

基于51单片机的电子鼻的设计与实现

研制了一套由51单片机和金属氧化物半导体气敏传感器阵列组成的便携式电子鼻系统,介绍了系统的工作原理和试验过程.对不同气体通过传感器阵列产生的响应信号进行了数据采集,并对各条响应曲线进行了特征分析,得出了适合单片机处理的特征向量.在低成本的51单片机中编制了神经网络识别程序,神经网络训练过程则在PC机上完成.在51单片机中用排序法和BP神经网络对样本进行分析识别,模拟人和动物的气味学习、记忆过程来学习气味特征、识别不同气味.文中阐述了详细的实验过程和部分气体特征向量,实际识别正确率为90%以上.     

Colorimetric Sensor Arrays System Based on FPGA for Image Recognition

A FPGA-based image recognition system is designed for eolorimetric sensor array in order to recognize a wide range of volatile organic compounds. The gas molecule is detected by the responsive sensor array and the responsive image is obtained. The image is decomposed to RGB color components using CMOS image sensor. An embedded image recognition archi- tecture based on Xilinx Spartan-3 FPGA is designed to implement the algorithms of image recognition. The algorithm of color coherence vector is discussed in detail compared with the algorithm of color histograms, and experimental results demonstrate that both of the two algorithms could be analyzed effectively to represent different volatile organic compounds according to their different responsive images in this system.     

A 128-pixel CMOS image sensor with integrated analog nonvolatilememory

A 128-pixel complementary metal-oxide-semiconductor (CMOS) image sensor array with analog nonvolatile storage for each pixel has been realized in a 1.5-μm single-poly standard CMOS/EEPROM technology and successfully tested. The integrated nonvolatile memory allows an offset correction for each sensor element, cancellation of the fixed pattern noise, and compensation of the background illumination. The sensor array can also learn a presented pattern and store it in its analog nonvolatile memory just by “seeing”. The stored pattern can be read out directly or, in combination with the optical input, it can be used for pattern recognition or motion detection. The required programming circuitry for the analog memory has been integrated on the same chip     

纳米ZnO基掺杂气敏元件阵列的制备与特性

以金属蒸气氧化法制备的纯纳米ZnO为气敏原料,通过丝网印刷技术在Al2O3基片上制得纯ZnO和掺杂ZnO的气敏元件阵列。结果表明,元件阵列具有低的功耗,纯ZnO气敏元件阵列在350~400℃对橙汁、可乐、酒精和汽油有较高的敏感性,灵敏度分别为2.9,2.9,53.5和43.4。通过Bi2O3+Cu2O的掺杂,可以降低纯ZnO的电导,并进一步提高气敏元件在250~350℃温度区间对汽油的敏感性。并对其气敏机理进行了探讨。