改进APEX算法在图像增强中的应用

APEX是一种基于快速傅立叶变换的直接盲解卷积算法,能够快速复原复杂纹理图像.该算法适用于平移不变且以2维放射对称Le'vy Stable概率密度函数(G类点扩散函数)作为卷积核的图像.论述了APEX算法的原理,并针对APEX在适应性和稳定性方面的不足提出了改进方法.实验表明,改进的APEX算法能够有效应用在多种真实模糊图像上,增强图像对比度和锐化图像细节,使图像的视觉效果得到明显改善.     

改进Mask RCNN的盾构隧道渗漏水检测方法

渗漏水是盾构隧道结构存在潜在损伤或缺陷的重要表征,快速、准确检测出渗漏水位置,对隧道安全运营和维护具有重要意义。现有的方法大多采用光学影像对隧道渗漏水进行检测,受隧道内空间和光线条件限制,难以获得高质量病害图片。因此,本文提出了一种基于激光点云数据与改进Mask RCNN相结合的渗漏水检测方法。首先对激光点云反射强度进行修正;然后生成灰度图像并建立渗漏水病害数据集;最后在Mask RCNN算法中引入空洞卷积和变形卷积,实现了隧道渗漏水病害的快速检测。利用某地铁采集的数据进行验证,结果表明,本文提出的改进Mask RCNN算法相较于原始算法和FCN算法检测精度均有明显提升,在盾构隧道渗漏水识别方面性能表现较好。     

改进的OPTA算法在遥感影像细化中的应用

在分析和研究快速细化算法和OPTA细化算法基础上,针对快速细化算法细化不彻底和OPTA算法模板设计的缺点提出了对OPTA细化算法的改进,设计了新的细化算法模板。经过实验证明改进的OPTA细化算法能够满足细化的基本要求,既保证了细化结果线条的单像素宽,又保持了原有图像线条的连通性,同时线条细节特征没有丢失,使细化结果得到了较大改善。     

遥感图像质量等级分类的深度卷积神经网络方法

遥感图像应用发展对图像质量的要求越来越高,不同质量的遥感图像往往需要不同的处理方法和参数。通过遥感图像质量等级分类研究,不仅能够为遥感图像的处理提供先验信息,还能够对遥感图像的客观质量评价和传感器的成像效果进行评估。为了克服现有的遥感图像质量等级分类方法计算参数获取困难、等级数量少的缺点,利用深度学习方法的分类机能,通过改进特征提取网络和等级分类设计,建立了一种基于深度卷积神经网络的遥感图像质量等级分类模型。通过质量等级分类预处理后,利用经典的深度学习方法进行目标检测实验。结果表明,所提方法在西北工业大学遥感图像数据集上质量等级分类的准确率、召回率、精确率和F1最高能达到0.976、0.972、0.974和0.973,优于传统算法。利用卷积神经网络实现遥感图像质量等级分类,既拓展了深度学习的应用领域,又为遥感图像质量评估提供了一个新方法。     

真正射纠正中地物遮挡区域自动检测与标识方法研究

本文基于真正射纠正的原理,探讨了图像遮挡区域处理的方法。针对真正射纠正中重叠区域和建筑物遮挡区域等影像的检测与标识问题,基于Z-Buffer算法和阈值选择法,通过将图像一分为四进行真正射纠正对Z-Buffer算法进行改进。试验表明,所提出的改进算法能够实现快速真正射纠正,具有良好的应用效果。     

级联卷积神经网络的遥感影像飞机目标检测

传统遥感影像飞机目标检测算法依赖于人工设计特征,对大范围复杂场景和多尺度的飞机目标稳健性较差,基于深层卷积神经网络的目标检测算法通常难以有效应对大幅影像的目标搜索和弱小目标检测问题,针对上述问题,本文提出了一种基于级联卷积神经网络的遥感影像飞机目标检测算法。首先根据全卷积神经网络能够支持输入任意大小图像的特点,采用小尺度浅层全卷积神经网络对整幅影像进行遍历和搜索,快速获取疑似飞机目标作为兴趣区域,然后利用较深层的卷积神经网络对兴趣区域进行更精确的目标分类与定位。为提高卷积神经网络对地物目标的辨识能力,在卷积层中引入多层感知器,并在训练过程中采取多任务学习与离线难分样本挖掘的策略;在测试阶段,建立影像金字塔进行多级搜索,并结合非极大值抑制消除冗余窗口,从而实现由粗到精的飞机目标检测与识别。对多个数据集下多种复杂场景的遥感影像进行测试,结果表明,本文方法具有较高的准确性和较强的稳健性,可为大幅遥感影像的飞机目标检测问题提供一个快速高效的解决方案。     

基于卷积运算的影像镶嵌算法研究

本文将距离加权运算运用到卷积运算上提出了改进的图像镶嵌算法。该算法针对重叠区域图像边缘灰度之间的差异,考虑到中心像元与周围像元之间的相关关系,即临近像元效应,因而需要采用一定的方法来消除或减小这种影响效应,基于空间分析方法给出了像元权重因子的求解,确定出了周围像元对中心像元灰度的影响。之后运用此权重因子采用卷积距离加权算法实现了影像的无缝镶嵌。文中最后对算法在时间和视觉效果上做出了一定的评价。经与其他3种镶嵌算法分析对比,该算法简单易行、耗时少、效率高,达到了良好的视觉效果。     

多尺度深度特征融合网络的遥感图像目标检测

本文针对现有方法对遥感图像目标检测准确率低的问题,在更快速区域卷积神经网络Faster R-CNN(Faster Region Convolutional Neural Networks)算法的基础上对其进行改进,提出一种新的遥感图像目标检测算法。该算法把Faster R-CNN算法中的VGG (Visual Geometry Group)特征提取网络替换为残差网络ResNet(Residual Networks),在此基础上加入特征金字塔网络以充分表达语义信息和位置信息,并使用焦点损失函数替代Faster R-CNN算法中的交叉熵损失函数以解决难易样本对总损失贡献的权重问题,最后对NWPU VHR-10数据集和RSOD数据集采用数据增广方法以解决数据集中图像样本数量少的问题。为验证本文算法的效果,进行了两组对比实验。第一组实验为本文提出的改进模块在NWPU VHR-10数据集和RSOD数据集上的消融实验;第二组实验为本文算法与其他算法在NWPU VHR-10数据集上的对比实验。实验结果表明,本文算法在NWPU VHR-10数据集和RSOD数据集上的多类平均准确率分别达到93.4%和9...     

卷积神经网络在高分遥感影像分类中的应用

针对目前应用于高分辨率遥感影像分类的常用算法,其精度已无法满足大数据环境下的分类要求的问题,该文提出了卷积神经网络分类算法。卷积神经网络模型降低了因图像平移、比例缩放、倾斜或者共他形式的变形而引起的误差。在大数据环境下,采用卷积神经网络算法对高分辨率遥感影像进行分类,避免了特征提取和分类过程中数据重建的复杂度,提高了分类精度。通过实验比对分析,证明了卷积神经网络在高分辨率遥感影像分类中的可行性及精度优势,对遥感图像处理领域等相关工作提供了参考价值。     

无人机采场图像二维经验小波变换降噪研究

针对传统图像降噪方法难以适用于露天矿无人机(UAV)图像降噪的问题,该文提出了一种基于经验小波变换(EWT)算法思想的露天矿UAV图像快速降噪方法。采用Littlewood-Paley小波算子检测提取UAV图像噪声特征,以提高阈值函数匹配精度;借助改进的自适应H阈值函数对UAV图像噪声进行计算,提高UAV图像边缘细节度;并通过二维经验小波变换逆重构,从而提高UAV图像纹理精度。实验结果表明,本文算法能够实现露天矿UAV图像的快速降噪,保留边缘细节和纹理特征,并具有较好的降噪效果。     

Object-based classification of hyperspectral data using Random Forest algorithm

This paper presents a new framework for object-based classification of high-resolution hyperspectral data. This multi-step framework is based on multi-resolution segmentation (MRS) and Random Forest classifier (RFC) algorithms. The first step is to determine of weights of the input features while using the object-based approach with MRS to processing such images. Given the high number of input features, an automatic method is needed for estimation of this parameter. Moreover, we used the Variable Importance (VI), one of the outputs of the RFC, to determine the importance of each image band. Then, based on this parameter and other required parameters, the image is segmented into some homogenous regions. Finally, the RFC is carried out based on the characteristics of segments for converting them into meaningful objects. The proposed method, as well as, the conventional pixel-based RFC and Support Vector Machine (SVM) method was applied to three different hyperspectral data-sets with various spectral and spatial characteristics. These data were acquired by the HyMap, the Airborne Prism Experiment (APEX), and the Compact Airborne Spectrographic Imager (CASI) hyperspectral sensors. The experimental results show that the proposed method is more consistent for land cover mapping in various areas. The overall classification accuracy (OA), obtained by the proposed method was 95.48, 86.57, and 84.29% for the HyMap, the APEX, and the CASI data-sets, respectively. Moreover, this method showed better efficiency in comparison to the spectral-based classifications because the OAs of the proposed method was 5.67 and 3.75% higher than the conventional RFC and SVM classifiers, respectively.     

Calibration facility for airborne imaging spectrometers

A new facility designed to perform calibration measurements of airborne imaging spectrometers was established at the German Aerospace Center (DLR) in Oberpfaffenhofen. This Calibration Home Base (CHB) is optimized to characterize radiometrically, spectrally, and geometrically the APEX (Airborne Prism Experiment) imaging spectrometer, which is currently being developed under the authority of the European Space Agency (ESA). It however can be used for other optical sensors as well. Computer control of major laboratory equipment allows automation of time consuming measurements. In APEX configuration (wavelength range: 380 to 2500 nm, instantaneous field of view: 0.48 mrad, field of view: ±14) spectral measurements can be performed to a wavelength uncertainty of ±0.15 nm, geometric measurements at increments of 0.0017 mrad across track and 0.0076 mrad along track, and radiometric measurements to an uncertainty of ±3% relative to national standard. The CHB can be adapted to similar sensors (including those with thermal infrared detectors) by exchanging the monochromator’s lamp, the gratings and the filters, and by adjusting the distance between the sensor and folding mirror.     

Improving radiometry of imaging spectrometers by using programmable spectral regions of interest

Programmable imaging spectrometers can be adjusted to fit specific application requirements that differ from the instrument initial spectral design goals. Sensor spectral characteristics and its signal-to-noise ratio (SNR) can be changed by applying customized online binning patterns.We present a software utility that generates application driven spectral binning patterns by using an SNR dependent sensor model. The utility, named BinGO (BInning patterN Generator and Optimiser), is used to produce predefined binning patterns that either (a) allow an existing imaging spectrometer to optimize its spectral characteristics for a specific application, (b) allow an existing imaging spectrometer to spectral and/or spatially emulate another instrument, or (c) design new multispectral or imaging spectrometer missions (i.e. spaceborne, airborne, terrestrial). We present a variety of BinGO case studies, including the simulation of airborne (APEX) [Itten, K.I. et al., 2008. APEX — The hyperspectral ESA Airborne Prism Experiment. Sensors 8(1), 1–25], spaceborne (SENTINEL III) [Nieke, J., Frerick, J., Stroede, J., Mavrocordatos, C., Berruti, B., 2008. Status of the optical payload and processor development of ESA’s Sentinel 3 mission. In: Proceedings of the Geoscience and Remote Sensing Symposium IGARSS 2008, pp. 427–430], as well as scientific and performance optimized approaches. We conclude that the presented approach can successfully be used to increase the efficiency of spectral information retrieval by using imaging spectroscopy data and to simulate various missions and requirements, finally supporting proper trade-off decisions to be made between performance optimization and scientific requirements. In addition, if specific sensor parameters are known, BinGO can also model other imaging spectrometers.     

Assessing the performance of two unsupervised dimensionality reduction techniques on hyperspectral APEX data for high resolution urban land-cover mapping

Despite the high richness of information content provided by airborne hyperspectral data, detailed urban land-cover mapping is still a challenging task. An important topic in hyperspectral remote sensing is the issue of high dimensionality, which is commonly addressed by dimensionality reduction techniques. While many studies focus on methodological developments in data reduction, less attention is paid to the assessment of the proposed methods in detailed urban hyperspectral land-cover mapping, using state-of-the-art image classification approaches. In this study we evaluate the potential of two unsupervised data reduction techniques, the Autoassociative Neural Network (AANN) and the BandClust method – the first a transformation based approach, the second a feature-selection based approach – for mapping of urban land cover at a high level of thematic detail, using an APEX 288-band hyperspectral dataset. Both methods were tested in combination with four state-of-the-art machine learning classifiers: Random Forest (RF), AdaBoost (ADB), the multiple layer perceptron (MLP), and support vector machines (SVM). When used in combination with a strong learner (MLP, SVM) BandClust produces classification accuracies similar to or higher than obtained with the full dataset, demonstrating the method’s capability of preserving critical spectral information, required for the classifier to successfully distinguish between the 22 urban land-cover classes defined in this study. In the AANN data reduction process, on the other hand, important spectral information seems to be compromised or lost, resulting in lower accuracies for three of the four classifiers tested. Detailed analysis of accuracies at class level confirms the superiority of the SVM/Bandclust combination for accurate urban land-cover mapping using a reduced hyperspectral dataset. This study also demonstrates the potential of the new APEX sensor data for detailed mapping of land cover in spatially and spectrally complex urban areas.     

利用GPS-IR监测土壤含水量的反演模型

GPS-IR（GPS-interferometric reflectometry）本质上是一种基于GPS辐射源的双基地雷达技术,利用大地测量型接收机记录的信噪比（signal-to-noise ratio,SNR）数据可用于反演土壤含水量。针对GPS-IR获取土壤含水量的参数估计问题,提出了一种改进的反射信号参数估计方法,并研究了土壤含水量反演模型的建立过程。实验结果表明,利用改进的反射信号参数估计方法可获得更加准确可靠的结果,反射信号相位与土壤含水量间存在显著的线性相关,可建立土壤含水量的线性反演模型,但在连续降雨条件下会存在较大误差。     

椭球面上图斑面积计算方法的改进

由于高斯投影存在面积变形,当面积精度要求较高时,需要在地球椭球面上进行图斑面积计算。通过对图斑椭球面积计算方法进行研究,改进了图斑面积的定积分近似计算方法,并利用改进的矩形法进行了实证分析。研究结果显示,与常规矩形法相比,改进的矩形法具有较高的计算效率,可以替代图斑椭球面积计算方法中的中间层算法和底层算法,简化图斑椭球面积的计算过程。借助C#语言的十进制数类型变量,利用改进算法获得了椭球面梯形面积的高精度可靠值。利用改进算法可以计算任意一大梯形图块的椭球面积,进而方便地计算任意图斑的椭球面积。     

南水北调工程某标段GPS高程线性拟合改进方法研究

基于GPS大地高不能直接应用于工程而需要转换为正常高的需求,针对南水北调工程某标段GPS控制线形特点,分别采用正交函数、三次样条与Akima曲线拟合及相应改进方法进行了GPS高程拟合实践应用研究,结果表明针对线性拟合区域在选点适宜的情况下,三种改进后的方法均能得到精度提升,其中采用改进后的Akima曲线拟合方法拟合结果的精度提高了近三倍,每千米水准测量全中误差达到了6.07 mm,显示了改进后的Akima曲线拟合方法针对此选点方案的优越性.      

改进的K-means算法在校车站点布局中的应用

校车站点布局问题是一种典型的设施选址问题.大多数设施选址问题均属于区域选址,将设施选址限定在一定空间范围内,其方法并不适用于需要将位置固定在道路上的校车站点布局问题.因此,为保证生成的站点分布在路网上,且学生到站点的步行距离较短,提出了一种改进的K-means聚类算法,以一定范围内密度最大的点为初始类中心,在逐次迭代中...     

基于GNSS TEC的电离层自相关预报误差分析及参数优化

电离层延迟是全球卫星导航系统(GNSS)高精度导航定位应用中的重要误差源.通过对电离层总电子含量(TEC)进行测量和短期预报可有效提升GNSS单频用户的定位精度,对其他无线电系统的电离层效应也可起到有效减缓作用.近二十年来提出了很多行之有效的短期预报方法,但还没有哪一种方法有绝对优势,其预测精度都有待提高.利用Madrigal数据库任意选取的5个格网点的TEC观测数据,首先比较了自相关和自回归滑动平均(ARIMA)方法,然后研究了自相关预报方法中实际参与加权的观测值覆盖天数和观测值数量这两个参数的选取对预报误差的影响,并提出了参数设置优化方案.试验结果显示：1)自相关方法的预报误差略小于ARIMA方法,且自相关方法所花费时间比ARIMA方法少,总体上自相关方法是一种性能更优的方法;2)对于自相关方法,相比于传统的“4+12”参数设置方案,“3+9”方案总体上具有更优的预报性能,说明TEC时间序列的当前状态可能主要与前3天的状态有关.相关结果可作为电离层短期预报工程实现的一个有用的参考方案.     

无人直升机面阵影像高精度对地定位

为提高无人机对地定位精度,必须对系统进行几何标定与补偿。首先进行了直接地理定位实验,检验系统在不经过任何标定时的直接地理定位精度;然后进行了自由网光束法平差实验,以验证无地面控制点时系统的定位精度;最后进行了POS辅助自检校光束法平差,对内方位元素变化、相机畸变、IMU安置角进行标定和补偿,并验证了系统误差补偿后的直接地理定位精度。实验结果表明,未经几何标定的无人直升机直接地理定位的精度可满足应急条件下测绘保障要求;少量地面控制点的辅助下定位精度远优于1∶500比例尺地形图测图的要求,可用于常规测绘生产;系统误差补偿后,直接地理定位精度有明显提升:平面精度提高46.37%,高程精度提高62.49%;系统误差标定和补偿方法正确、有效,无人直升机具有很好的测绘应用前景。