改进的基于神经网络的卫星遥感图像去条带方法

由于焦平面阵列的非均匀性影响,线阵列推扫所成的遥感图像表现出规律的条带失真,严重地影响图像质量。本文介绍一种基于神经网络的条带消除方法,以及利用该方法对实际遥感图像校正的结果。     

伺服电机及系统特性性能测试系统设计

介绍了用于伺服电机及系统特性性能试验的测试系统。该系统采用能量回馈的方式进行对拖加载试验,能够满足0.75~110 kW伺服电机及控制器的交/直流伺服电机系统的性能测试。     

FAS100氧分析仪应用于氮气中微量氧分析

阐述了FAS100氧分析仪在氮气微量氧分析中的应用,分别说明了FAS100仪器与其他氧分析仪的区别和该仪器的优点.     

1．6GHz～2．8GHz宽带高增益低噪声放大器

介绍了1.6 GHz～2.8 GHz宽带高增益低噪声放大器的研制工作.采用微波宽带匹配和CAD技术,利用网络分析的方法,研制出了符合整机要求的微波放大器.其主要性能指标为工作频率1.6 GHz～2.8 GHz,增益≥50 dB,增益平坦度≤±1 dB,噪声系数≤2 dB.     

An approach to multi-temporal MODIS image analysis using image classification and segmentation

This paper discusses an approach for river mapping and flood evaluation based on multi-temporal time series analysis of satellite images utilizing pixel spectral information for image classification and region-based segmentation for extracting water-covered regions. Analysis of MODIS satellite images is applied in three stages: before flood, during flood and after flood. Water regions are extracted from the MODIS images using image classification (based on spectral information) and image segmentation (based on spatial information). Multi-temporal MODIS images from “normal” (non-flood) and flood time-periods are processed in two steps. In the first step, image classifiers such as Support Vector Machines (SVM) and Artificial Neural Networks (ANN) separate the image pixels into water and non-water groups based on their spectral features. The classified image is then segmented using spatial features of the water pixels to remove the misclassified water. From the results obtained, we evaluate the performance of the method and conclude that the use of image classification (SVM and ANN) and region-based image segmentation is an accurate and reliable approach for the extraction of water-covered regions.     

Correlation between the cosmic noise absorption calculated from the SARINET data and the energetic particles measured by MEPED: Simultaneous observations over SAMA region

The cosmic noise absorption is presented in terms of two-dimensional images obtained from the imaging riometers operated at the Southern Space Observatory (geographic coordinate: 29.4° S, 53.1° W), in São Martinho da Serra, Brazil, Concepcion (geographic coordinate: 36.5° S, 73.0° W) and Punta Arenas (geographic coordinate: 53.0° S, 70.5° W) in Chile, which belong to the South American Riometer Network and are located at the central and periphery regions of the South American Magnetic Anomaly. Correlations are performed between the maximum cosmic noise absorption observed at these stations and the energetic electron flux in two energy channels (>30 and >300 keV) and the proton flux in three energy channels (80–240, 800–2500 and >6900 keV) as measured by the Medium Energy Proton and Electron Detector, during a moderate geomagnetic storm that occurred on September 3, 2008. The results show high correlations between the cosmic noise absorption detected at São Martinho da Serra and the flux of protons with energy between 80 and 240 keV, and the flux of electrons with energies higher than 300 keV, while an additional ionization at Concepcion was correlated with electrons of energies higher than 30 keV. The cosmic noise absorption detected at Punta Arenas was probably caused by the increase of the protons flux with energy between 80 and 240 keV.     

基于拉普拉斯矩阵的网络社区划分算法在电缆网测试中的应用

在低频电缆网导通绝缘测试中,对电缆网连接节点的划分是开展低频电缆网测试的基础。通过电缆网连接节点的划分,可以高效低成本完成转接电缆和测试方案的设计,将能够大大节省电缆网性能测试的周期和成本,为工程项目任务中电缆网快速、高效、准确测试奠定基础。目前,对电缆网的测试一般直接通过连接器进行转接电缆设计,这种方法由于转接电缆连接器芯点分配的问题,导致测试效率较低。针对以上问题,提出改进型的转接电缆设计方法,并针对改进型设计方法中子网络难划分的痛点问题,采用基于拉普拉斯矩阵的网络社区划分算法完成电缆网相关性大的子网络的划分。该方法解决了节点多、关联性强的电缆网子网络划分问题,经过仿真验证,使用划分后设计的转接电缆进行电缆网测试,测试效率得到提升。     

基于神经网络自适应稳定PID控制方法的研究

经典的基于对象精确数学模型的PID控制方法的自适应性较差,难以适应具有非线性、时变不确定性的被控对象.神经网络控制算法的稳定性又受到迭代初值的影响,且算法复杂.为此提出了一种基于RBF神经网络的、结构简单的、稳定的PID直接自适应控制方法.讨论了控制器参数迭代初值选取的基本原则,并给出了在保证系统稳定性前提下参数的迭代算法.仿真研究结果表明,该方法的鲁棒性和跟踪性能均优于经典PID方法.      

CPMIS综合查询及数据维护技术方法与功能分析

通过对产品管理信息系统工作情况的分析,介绍了实现综合查询及数据维护部分的技术方法,并对其功能进行了具体的剖析。     

基于全连接神经网络的分层旋流火焰燃烧振荡预报

为了指导主动控制系统抑制燃烧振荡,有必要针对不同的燃烧振荡预报手段开展研究和验证.以甲烷预混同心分层旋流火焰的时均图像为基础,采用降低图像分辨率和提取火焰结构特征参数这两种不同的方式对火焰图像信息进行简化处理,并使用全连接神经网络对燃烧振荡进行预报研究.结果发现,两种方式都可以较为准确地预报燃烧振荡,精度均达到90％以...