多特征融合的高分辨率影像建筑物变化检测

为充分发挥遥感影像中各特征的优势,提高遥感影像建筑物变化检测精度,基于面向对象的分析方法,提出了一种基于模糊集合的证据理论特征信息融合的变化检测方法。首先,在影像分割的基础上,利用变化矢量分析法分别计算前后时相对应对象的光谱、纹理特征差异及形态学建筑物指数差异;然后,以Sigmoid函数作为隶属度函数,计算对象属于变化类和非变化类的隶属度并以之构建证据理论所需的基本概率分配函数;最后,利用证据理论对多种特征进行融合并通过规则判定得到建筑物变化区域。利用不同地区影像的试验结果表明,该方法能够有效融合影像的多种特征,提高建筑物变化检测的精度。     

半监督离散势理论在遥感影像变化检测中的应用

随着遥感技术的发展,遥感影像变化检测作为一种有效的技术手段,在环境监测、灾害救援等领域发挥了重要作用。然而地物复杂、标记困难等问题导致有效的变化检测存在一定的困难。本文提出了一种基于半监督离散势理论的遥感影像变化检测方法。该方法首先采用一种新的标记样本点的方法得到训练集,然后利用KNN方法构造复杂网络,最后对复杂网络中经典Wu-Huberman算法进行改进并划分网络。所得到的两个社团结构恰好对应了变化部分和不变部分。试验结果表明,基于半监督离散势理论的变化检测方法具有良好的变化检测性能。     

网络三维数字城市海量模型优化算法研究

大规模海量三维模型数据网络传输和终端可视化是面向Web的三维GIS应用关键技术。本文提出将三维模型自适应面删除简化算法和地形模型单面片自动生成算法相结合,实现大批量构建面简化LOD （层次细节模型）模型库自动建库,有效解决了海量高精度、大范围三维模型面向Web的三维GIS应用问题。研究成果在天津市网络三维数字城市规划管理系统项目中成功进行试验,效果显著。     

Mapping shoreline indicators on a sandy beach with supervised edge detection of soil moisture differences

This study describes a method to map shoreline indicators on a sandy beach. The hypothesis is that, on this beach, spectral albedo is predominantly determined by moisture content and water lines can, therefore, be detected as albedo contrasts. A laboratory experiment is performed to relate moisture content to image albedo, and supervised edge detection is subsequently used to map the shoreline indicators with remote sensing imagery. The algorithm is tested with data from visible, near-infrared and shortwave-infrared wavelength regions. These results are compared to shoreline indicators obtained by a field survey and a shoreline indicator derived from a digital elevation model. Both the water line present when the imagery was acquired, as well as the maximum extent of the last flood, can be detected as a single edge. Older high water lines are confused with the last high water line and appear dispersed, as there are multiple debris lines present on the beach. The low water line, usually in saturated sand, also appears dispersed due to the presence of channels and troughs. Shorelines are constant moving boundaries, which is why shoreline indicators are used as a proxy. Unlike a mathematical indicator that is based on an elevation model, our method is more sensitive to the dynamic nature of shorelines. Supervised edge-detection is a technique for generating reproducible measurements of shoreline indicator positions over time, and aids in the monitoring of coastline migration.     

一种改进均值的自适应中值滤波算法

针对自适应中值滤波算法在滤除高浓度椒盐噪声和保留图像边缘细节中的不足,提出了一种改进均值的自适应中值滤波(IMAMF)算法。该算法采用扩充图像边界的方式,使得原图像的边界点能在自适应的滤波窗口下参与噪声检测和滤波处理,并在检测噪声和信号时,增加了噪声阈值判定,将存在噪声的像素点用修正后的均值滤波器值输出,信号点则用原始灰度值输出。为了验证算法的可行性,采用了5种不同的算法进行仿真对比分析,并从主观角度和客观指标上进行效果评价。试验结果表明:该算法能有效滤除浓度为10%～90%范围内的椒盐噪声,且图像细节和边缘信息得到了更好的保留,滤波性能明显优于其他算法。     

基于随机森林的FY-2G云检测方法

根据遥感影像中云检测原理,提出了基于随机森林的遥感影像云检测方法,并将其应用于FY-2G影像。结合国家气象卫星中心（NSMC）的云检测产品数据进行了算法的精度检验,云检测个例的精度检验结果显,最高命中率（POD）为88.32%,最低误报率（FAR）为9.36%,临界成功指数（CSI）为80.14%。结果表明,该方法有效地提高了云检测精度,同时能正确标识NSMC中部分误判的情况。     

一种适用于航空影像的无参考模糊探测方法

针对侧风、强风、湍流等飞行环境容易造成航空影像运动模糊,严重影响航空影像质量,同时航空影像数据量大,手动挑选模糊影像费时费力的问题,为了提高航摄内业人员的工作效率,该文研究一种适用于航空影像的自动模糊探测方法,以主流的无参考再模糊算法Reblur和无参考结构清晰度算法NRSS为基础,结合航空影像具有丰富地物的特点,对影像进行分块处理,计算所有字块的Reblur和NRSS模糊探测值,最后得到整幅影像的模糊探测值。其中,再模糊算法通过计算待测影像和参考影像的水平和垂直运动方向上的灰度变化来评价图像模糊度;NRSS算法在结构相似度SSIM算法基础上加入梯度信息提取和高斯滤波等改进,通过计算结构相似度评价图像模糊度。实验结果表明,该文研究的无参考模糊影像探测方法适用于航空影像数据,其评价结果与人眼主观评价结果具有较高的一致性,能够准确地缩小模糊影像的查找范围,极大地提高了航摄内业效率。     

An assessment of the impact of urbanization and land use changes in the fast-growing cities of Saudi Arabia

Understanding rates, patterns and types of land use and land cover (LULC) changes are essential for various decision-making processes. This study quantified LULC changes and the effect of urban expansion in three Saudi Arabian cities: Riyadh, Jeddah and Makkah using Landsat images of 1985, 2000 and 2014. Seasonal change of vegetation cover was conducted using normalised difference vegetation index, and object-based image analysis was used to classify the LULC changes. The overall accuracies of the classified maps ranged from 84 to 95%, which indicated sufficiently robust results. Urban area was the most changed land cover, and most of the converted land to urban was from bare soil. The seasonal analysis showed that the change of vegetation cover was not constant due to climatic conditions in these areas. The agricultural lands were significantly decreased between 1985 and 2014, and most of these lands were changed to bare soil due to dwindling groundwater resources.     

Normalizing shadows in multi-temporal aerial frame imagery using relative radiometric adjustments to support near-real-time change detection

This study addresses the problem of shadows in multi-temporal imagery, which is a key issue with change detection approaches based on image comparison. We apply image-to-image radiometric normalizations including histogram matching (HM), mean-variance (MV) equalization, linear regression based on pseudo-invariant features (PIF-LR), and radiometric control sets (RCS) representing high- and low-reflectance extrema, for the novel purpose of normalizing brightness of transient shadows in high spatial resolution, bi-temporal, aerial frame image sets. Efficient shadow normalization is integral to remote sensing procedures that support disaster response efforts in a near-real-time fashion, including repeat station image (RSI) capture, wireless data transfer, shadow detection (as precursor to shadow normalization), and change detection based on image differencing and visual interpretation. We apply the normalization techniques to imagery of suburban scenes containing shadowed materials of varied spectral reflectance characteristics, whereby intensity (average of red, green, and blue spectral band values) under fully illuminated conditions is known from counterpart reference images (time-1 versus time-2). We evaluate the normalization results using stratified random pixel samples within transient shadows, considering central tendency and variance of differences in intensity relative to the unnormalized images. Overall, MV equalization yielded superior results in our tests, reducing the radiometric effects of shadowing by more than 85 percent. The HM and PIF-LR approaches showed slightly lower performance than MV, while the RCS approach proved unreliable among scenes and among stratified intensity levels. We qualitatively evaluate a shadow normalization based on MV equalization, describing its utility and limitations when applied in change detection. Application of image-to-image radiometric normalization for brightening shadowed areas in multi-temporal imagery in this study proved efficient and effective to support change detection.     

融合颜色和深度信息的运动目标提取方法

行人检测是计算机视觉、智能交通等领域研究的热点与难点,基于深度传感器对室内复杂场景下的行人检测展开研究。目前,基于颜色与深度数据的目标检测方法主要包括基于背景学习的方法和基于特征检测算子的方法,前者依赖于视频序列头几十帧的背景知识,帧的数量决定检测质量;后者存在计算量大的问题,训练样本的不足也会影响行人检测结果。因此,深入分析了复杂场景特征,融合颜色和深度信息,提出了RGBD+ViBe(visual background extractor)背景剔除方法,实现前景运动目标的准确提取。实验结果表明,提出的RGBD+ViBe方法在前景运动目标检测准确率方面要明显高于仅考虑颜色或深度信息方法以及RGBD+MoG(model of Gaussian)方法。