Aircraft Auto-detection in Domestic High Resolution Remote SensingImages Using Deep-learning

It is of great significance to automatically detect aircrafts from remote sensing imagery to get their locations.However,due to aircraft posture variance,complicated background and incomplete outlines,it is challenging to achieve a high aircraft detection accuracy.Traditional aircraft detection methods are usually based on hand\|crafted features and machine learning based classifiers,which is not robust enough for the translation and rotation variations.To tackle the above issues,this paper introduces deep convolutional neural network and the strategy of transfer learning to detect aircrafts from Chinses domestic satellite remote sensing images.Specifically,this paper first constructs an aircraft sample database,which consists aircrafts of different sizes and poses.Afterwards,YOLO V2 trained with natural images is utilized as the detection model and is further fine\|tuned with aircraft samples to increase the robustness and performance.Experiments were done on the Shanghai Pudong airport from Chinese GF\|2 remote sensing data.Experimental results showed a good performance with a recall of 92.25% and a precision of 94.93%.It is indicated that deep learning together with model transfer can get a high aircraft detection accuracy with limited training samples.The method in this paper can be generalized to other land object detection problems which shows a good promotional value.     

Water Depth Information Extraction with Multi-spectralRemote Sensing in Coral Reefs Region

Water depth inversion is of great important to coral reefs’ protection and engineering construction of the coral reef region.As the correlation between the remote sensing radiance and the water depth is a very well,so the band ratio transform and the linear combination transform were adopted to the water depth inversion around coral reefs.Based on the complex topography of coral reefs,these two algorithms were optimized for the water depth inversion.Finally,our experimental results show that the improved ratio transform was suitable for retrieving water depths among 3~5 m and 5~10 m.While none of the models that mentioned in this paper performed well in water depth of 5~10 m,the preliminary conclusion was that coral reefs’ special topographic caused it.     

Radiometric Consistency between Landsat 8 OLIand Sentinel-2 MSI Imagery in Mountainous Terrain

The combined use of multi\|sensor/multi\|temporal images provides more opportunities for long\|term land surface monitoring with high resolution and frequency requirements.However,as sensors differ in their orbital,spatial,or spectral configuration,uncertainty was introduced in the radiometric consistency of multi\|sourse images,and that becomes more outstanding in mountainous terrain with the sharp topographic relief.Therefore,a series of radiometric corrections need to be carry out before further application.The objective of this study was to indicate the radiometric consistency of Landsat\|8 OLI and Sentinel\|2 MSI images.Thus the radiometric differences between the corresponding bands of these two images acquired almost simultaneously by OLI and MSI over 2 areas at different latitude was calculated for the TOA reflectance images first.Then several radiometric corrections(atmospheric correction,BRDF correction and bandpass adjustment) were carried out successively and after each of them the radiometric differences were researched again to assess the performance of each correction method.The results first indicate that there is high radiometric consistency between OLI\|L1T and MSI\|L1C images with the R²greater than 0.9 for each band involved.Then higher consistency was found after the 6S atmospheric correction and C\|factor BRDF correction,while no remarkable improve was found after the fixed\|parameter bandpass adjustment.Furthermore,in area with great topographic relief,the radiometric consistency were higher for hillside facing the sun than hillside in shadow (the MAD of SWIR2 band was 0.010 and RMSD was 0.007 in sun\|light area,while the MAD was 0.005 and RMSD was 0.004 in shadowed area).The results point out that proper atmospheric correction,BRDF correction and bandpass adjustment could be used to improve the radiometric consistency,and topographic correction might also be carried out to balance the radiometric consistency differences between different hillsides.     

A New Ship Target Detection Algorithm based on SVMin High Resolution SAR Images

The characteristics of ocean background and target in the high resolution synthetic aperture radar (SAR) images are analyzed.Aiming at the requirements of ship detection in high-resolution synthetic aperture radar (SAR) image,the detection accuracy,intelligence level,real-time and processing efficiency,we put forward a high resolution SAR images ship detection algorithm based on support vector machine.The algorithm designs a pre-training support vector machine (SVM) classifier and complete the screening of the ship target block area,then the algorithm of optimal entropy thresholds proposed by Kapur,Sahoo,Wong(KSW) will be used on the target area selected for fine detection of ship targets.In this paper,several commercial satellite data,such as TerraSAR-X,are used to verify the experiment.Comparing with the classical CFAR detection algorithm,Experimental results show that the algorithm can improve the false alarm caused by the speckle noise and ocean clutter background inhomogeneity.At the same time,the detection speed is also increased by 20% to 35%.     

大数据查询综述

大数据具有传统数据所不具有的数量大、种类多、速度快、真实性等特点,传统的数据查询技术满足不了日益增长的大数据查询需求,大数据查询技术应运而生并迅速发展。从大数据查询的角度出发,着重分析大数据存储技术、大数据处理平台以及大数据查询引擎等。分别对比介绍传统关系型数据库、NoSQL、NewSQL和它们在大数据查询处理上的应用,介绍当前流行的大数据处理平台以及在这些平台上运行的大数据查询引擎,对其优缺点进行了综合阐述。      

文本情感倾向性分析

Aiming at the problem of manual annotation in the text sentiment analysis, a new method based on five tuple of appraisal expression is proposed. This  method just needs appropriate sentiment dictionary. The sentiment tendencies of comments are analyzed without lots of markup work. Through the combination of unsupervised and supervised learning methods to construct the evaluation thesaurus and evaluation object list; the extraction of appraisal expression is based on these lists, using linear chain conditional random fields model, which is in the chain of sentiment words. Finally, evaluation objects are divided into four categories and emotional words are divided into five types according to the relationship between semantic collocation, combined with the influence of sentence pattern, negative word and degree word on the sentiment analysis, a method of calculating the sentiment tendency of the text is put forward. Compared with other methods, this method based on the appraisal expression has obtained better F value, and it has a certain cross domain.     

巴尔干战火纷飞互联网硝烟弥漫

科索沃引发信息战 由科索沃危机引发的南联盟地区的战火,已为20世纪的人类历史抹上了最后一道丑陋悲惨的色彩.而巴尔干这个曾经点燃过第二次世纪大战导火线的地区,也因为这次绝对称得上"灾难"的局势,再度成为世界关注的焦点.然而与以前战争有所不同的是,在这次战争中除了广大传统媒体的竞相报道,从而使得世界人们及时了解到战争的动向外,作为正迅猛发展的网络则成为战事的即时映像,令万里之外的人们也能对战局洞若观火.     

Influence of Sea Surface Wind Vector on Sea SurfaceBrightness Temperature Gain in Different Seasons

Based on the correlation of the sea surface wind vectors to the sea surface roughness temperature in different seasons,wind field data of the Windsat L2 U10 wind fields in the northwest Pacific in January,April,July and October from 2012 to 2016 were selected,and used sea surface roughness Semi\|empirical and theoretical algorithm,the relationship between the brightness temperature gain caused by wind speed and wind vector under different seasons was analyzed.the results showed that the contribution of wind speed to horizontal brightness temperature gain was greater than that of vertical brightness temperature; the change of horizontal brightness temperature gain was the most significant in January and the least was in July; the maximum and minimum mean values of wind speed to vertical brightness gain were 0.19 K and 0.05 K respectively,indicating that the wind speed had little effect on the vertical brightness gain.It showed that the vertical brightness temperature gain was almost independent of the seasons by the standard deviation calculation; in April and October to form larger cyclone phenomenon in high latitude regions by the Pacific and Hawaii high pressure under the influence,and with brightness gain changes,showing obvious features of the North Pacific gyre.     

Extraction and Analysis of Hyperspectral Data and Characteristics from Pedicularis on Bayanbulak Grassland in Xinjiang

Through the field acquisition of three vegetation spectral datas,flowering Pedicularis,non flowering Pedicularis and common vegetation on Bayanbulak grassland,the first derivative,the two derivative and the reciprocal logarithm transformation were used to the smoothed and denoised data to analyze the difference sensitive bands of vegetation.The results showed that in visible light,non flowering Pedicularis and common vegetation showed the overall consistency,but the spectral curve of flowering Pedicularis showed a significant difference.In the red band and near infrared band at 750nm,non flowering Pedicularis reflectance increased significantly,and the three kinds of spectral reflectance showed significant differences.The reciprocal logarithmic transformation in the visible 580~680 nm band could be used to distinguish the Pedicularis as sensitive area.The spectral reflectance difference between the three at 655 nm was the most obvious.That solved the non flowering Pedicularis and common vegetation confusable problems.The improved normalized difference vegetation index by calculation,to further validate and showed the reciprocal logarithmic transformed values NDVI RLR could be distinguished the difference of flowering Pedicularis,non flowering Pedicularis and common vegetation.The extraction and analysis of hyperspectral data and characteristics from Pedicularis provided a theoretical basis for remote sensing monitoring of Pedicularis,and Remote sensing technology has great significance in Pedicularis resource survey and monitoring application.     

天气雷达网络的进展

天气雷达在现代气象综合观测系统中起着非常重要的作用,雷达网络能够覆盖大的空间范围,有利于雷达间的数据融合,可以更为精细地观测不同尺度的天气过程。概述了天气雷达网络技术及其应用进展,包括收发分置双(多)基地雷达网络,美国的NEXRAD和TDWR网络、欧洲的OPERA计划和中国的CINRAD网络,最后介绍最新发展的CASA网络雷达系统。较为详尽地介绍CASA网络的端对端系统、DCAS模式和数据产品,并与NEXRAD的性能进行比较。CASA能够观测中小尺度强对流系统更细微的变化,同时也是S/C波段雷达网络的有益补充。S/C波段大功率长程雷达网络与X波段小功率短程雷达网络的结合及其资料的相互     

基于Android的实验室管理系统

通过对Android、JSON和WCF等相关技术进行研究,设计一款利用Android调用WCF服务模式的实验室管理系统。从系统的架构模式设计、功能模块设计、数据传输处理等方面进行详细介绍。通过软件测试,验证了该方案具有可行性。     

一种新的SAR图像斜距多普勒定位模型的直接解法

斜距多普勒定位模型是实现SAR图像无控制点定位的基本模型,因此如何高效准确地获得RD方程组的解成为实现SAR图像无控制点几何定位的根本问题,也是实现SAR自动图像定位的关键。首先利用地固坐标系下目标速度为零的特性,简化RD模型并整理为一元四次方程,然后利用一元四次方程求根方法,推导出斜距多普勒定位模型的明确数学解析解,并与常用的数值迭代求解方法进行比较分析,仿真结果验证了推导的正确性。新解法具有无需设置初始值,易于编程实现,鲁棒性高的特点。     

Estimation of Forest Aboveground Biomass and Canopy Heightin Jiangxi Province Using GLAS and Landsat TM Images

Aiming at the problem that optical remote sensing cannot estimate forest biomass exactly because it’s easily affected by the weather and hard to penetrate the canopy of the forest.Using Jiangxi forest as the study area,established forest canopy height and forest biomass model by GLAS waveform data,integrating multispectral data(TM) and filed survey data.The study results show:(1) using waveform feature parameter,terrain feature parameters and field survey data to build forest canopy height model can eliminate the terrain influence and obtain the discrete canopy height.(2) Combined with the NDVI and discrete canopy height can be carried out large scale continuous forest canopy height mapping.(3) Power function relationship between canopy height and forest biomass can be used to estimate forest biomass.In general,large\|footprint LiDAR combined with optical Landsat TM data can give full paly to the advantages of multi\|source remote sensing and improve the precision of forest biomass inversion.     

基于非下采样Contourlet变换域

依据非下采样Contourlet分解系数与其父系数之间的相关性,给出非高斯双变量分布模型,并基于该模型提出一种新的非下采样Contourlet变换图像分割方法。用合成纹理图像和实际图像进行仿真实验,并与小波域隐马尔可夫树模型分割及Contourlet域隐马尔可夫树模型分割等方法进行了比较,实验结果表明,在大多数情况下,该算法分割结果要好于相比较的方法,在边缘特征方面保持了良好的视觉效果,并且模型的训练简单快速。     

青海省土地荒漠化遥感动态监测

近年来由于气候变化和人类活动的影响,青海省受荒漠化的影响日趋严重,采取有效的手段进行荒漠化快速动态监测,跟踪荒漠化现状和变化趋势,是荒漠化防治必要手段。基于像元二分模型,采用2000~2012年MODIS的增强型植被指数产品(EVI)估算青海省植被覆盖度。以植被覆盖度为主要评价指标,对青海省荒漠化现状和动态变化进行快速监测。结果表明:遥感估算的荒漠化面积和全国荒漠化调查的结论基本一致。13年间,青海省荒漠化土地的空间分布特征没有明显变化,全省荒漠化和潜在荒漠化土地总面积在研究时段内没有明显变化趋势,但是受自然因素影响,面积年际波动明显。重度荒漠化面积整体上呈明显下降趋势,青海省沙化和荒漠化状况有所改善,部分地区生态群落生产力、草地植被覆盖度呈上升趋势,表明近年来在青海实施的林业工程和荒漠化治理措施逐步显现出生态效益,发挥了实际作用。     

面向对象的土地覆被变化检测研究

运用面向对象的方法进行土地覆被变化检测,利用遥感数据光谱信息、纹理特征、拓扑关系,在多尺度分割获得对象的基础上,构建了变化矢量方法和向量相似性的检测方法,两种检测方法均成功检测出了所选取实验区的土地覆被变化信息。结果表明:对于同一区域同一时相的两期影像的面向对象变化检测,两种方法的总体精度都在80%以上,但变化矢量方法(CVA)精度要高于向量相似性方法。因此,在进行土地覆被变化检测时可以优先考虑变化矢量方法(CVA)。     

水体后向散射特性研究进展

后向散射系数是水体重要的固有光学量之一,是水色反演半分析模型的重要输入参数,与水体组分有着密切关系。回顾了近年来国内外有关水体后向散射特性的研究进展,主要包括纯水、浮游植物和悬浮泥沙等无机颗粒物的后向散射特性,以及不同颗粒物的后向散射光谱变化性等。此外,还阐述了后向散射系数的理论基础和测量方法,并探讨了影响水体后向散射特性的影响因子,最后提出了水体后向散射的研究方向和发展前景。     

非圆传播算子DOA估计算法

传播算子方法(Propagator method,PM)是一种常用的波达方向估计算法,但传统PM算法在阵元数较小时估计精度较低.本文针对非圆信号,提出了一种非圆传播算子DOA估计算法.该算法首先对接收信号进行扩展,而后对扩展信号进行子阵划分,最后利用PM方法获得非圆传播算子.算法将信号的非圆特性与传统PM算法相结合,既具有传统PM算法无需计算协方差函数的特点,又去除了峰值搜索过程,减小了算法运算量.同时非圆特性的引入,使算法估计精度有了大幅度提高.通过计算量分析说明了算法的计算复杂度较小,并通过仿真验证了算法的有效性.