Classification-based vehicle detection in high-resolution satellite images

In this study, we have looked into the problem of vehicle detection in high-resolution satellite images. Based on the input from the local road authorities, we have focused not only on highways, but also on inner city roads, where more clutter is expected. The study site is the city of Oslo, Norway. To do vehicle detection in these areas, we propose an automatic approach, consisting of a segmentation step, followed by two stages of object classification. In the process, we utilize multispectral images, panchromatic images and a road network. The approach has been tested on Quickbird images, and the results that are obtained have been compared with manual counts and classifications.     

Road detection from high-resolution satellite images using artificial neural networks

This article treats the possibility of using artificial neural networks for road detection from high-resolution satellite images on a part of RGB Ikonos and Quick-Bird images from Kish Island and Bushehr Harbor, respectively. Attempts are also made to verify the impacts of different input parameters on network's ability to find out optimum input vector for the problem. A variety of network structures with different iteration times are used to determine the best network structure and termination condition in training stage.It was found that when the input parameters are made up of spectral information and distances of pixels to road mean vector in a 3 × 3 window, the network's ability in both road and background detection can be improved in comparison with simple networks that simply use spectral information of a single pixel in their input vector.     

多核多类关联向量机的高分辨率影像目标检测

从高分辨率遥感图像数据中准确检测多类目标的任务对于检测速度和模型训练时间提出了较高的要求.文章提出了一种MKL_mRVM方法:该方法采用基于快速边缘似然最大算法直接计算mRVM分类器的决策函数,避免了传统RVM重复计算目标函数Hessian矩阵的过程,并且因为不需要构造一系列两类分类器,缩短了多类模型的训练时间;同时,将多个基础核引入多类模型,训练过程中采用交叉验证方法确定基础核权重,在随机分出的确认集上检验分类器的精度,选取使得分类模型精度最高的值作为权重的优化结果.实验结果表明,该方法能够在保持解的稀疏性的前提下,有效地缩短模型训练时间.     

融合多特征的遥感影像变化检测

常规多时相遥感影像变化检测主要基于光谱信息,没有充分利用纹理、几何、形状等多种特征信息,不足以体现检测目标的完整性和准确性。本文针对不同特征在变化检测中应用的优势,在提取影像多种特征的基础上,构建了1维和多维两种基于信息融合策略的变化检测方法,即利用1维特征空间加权距离相似度运算、多维特征空间的模糊集融合和支持向量机融合策略进行变化检测。利用多时相QuickBird高分辨率遥感影像进行城市土地覆盖变化检测试验,结果表明,本文方法可以有效集成不同特征的优势与表征变化信息的能力,提高变化检测过程的稳定性和适用性,同时能够更好地保持变化地物的结构和形状,突出主要变化目标。     

矢量数据辅助的高分辨率遥感影像道路自动提取

高分辨率遥感影像上细节信息繁杂、干扰物普遍存在,对其进行自动化道路识别与提取的相关研究仍处在探索阶段。在道路提取过程中引入矢量数据辅助,可解决初始信息获取的困难,得到可靠性较强的训练样本。为此,提出一种矢量数据辅助下的道路提取方法,能够筛选出矢量数据中包含的有效信息,引导实现对高分辨率遥感影像的道路自动提取。利用Mean-shift滤波对图像进行预处理后,首先从矢量数据获取候选种子点,并通过提炼同质区域的形状特征剔除错误候选点;然后,自动获取负样本点以进行朴素贝叶斯分类,并采用邻域质心投票算法从分类影像提取道路中心线;最后,结合像素跟踪与方向判断矢量化道路中心线,并提出一种基于矢量几何分析的断线连接与毛刺剔除方法,对提取结果进行信息修复与规整、优化。实验结果显示,该算法的提取质量达到80%以上,且具备较强的稳健性,能够适应具有不同道路辐射和分布特征的高分辨率遥感影像。     

国产高分辨率卫星影像云检测方法分析

云检测方法大都针对特定的传感器或依赖多个波段,对参数要求高,而国产高分辨率卫星影像通常包含波段数较少,多数云检测方法不适用。本文采用深度学习的方法,以融合后的高分一号影像为例,应用基于双重视觉注意机制模型进行云检测,并与人工采集、全卷积网络模型的检测结果进行对比。理论分析和研究结果表明:基于双重视觉注意机制的模型云检测结果与人工采集进行对比,正确率为0.986 4;通过增加云样本数量和非云样本数量可有效解决模型对道路、河流、居民地的误检测问题;基于双重视觉注意机制的模型与全卷积网络模型相比,云边界更为准确,模型适用性更强。利用较少的波段信息进行云检测为国产其他高分辨率卫星影像云检测提供了参考。     

多尺度分割的高分辨率遥感影像变化检测

针对高空间分辨率的遥感影像,提出了一种基于多尺度分割的变化检测算法。采用Mean-Shift分割算法对影像进行多尺度分割,构建了不同尺度上的地理对象,以不同尺度上的地理对象灰度均值构建了变化检测的多尺度特征向量,采用变化矢量分析法获得最后的变化检测结果。以城镇区和农田区的Quick Bird影像对本文算法进行了检验,从精度评价的效果来看,无论城镇区还是农田区,采用面向对象的变化检测方法精度都高于基于单像素的检测方法,且当尺度层数固定时,多尺度组合的变化检测结果优于单一尺度的变化检测结果,对城镇、农田区域的变化检测的精度分别达到87.57%和81.55%。本文算法既可以顾及大面积同质区域变化,又可以反映小的地物目标及边缘部分的变化,能够很好地满足城镇、农田等不同环境背景下的变化检测需求,在国土资源监测中具有一定的应用价值。     

改进支持向量机的高分遥感影像道路提取

针对支持向量机受分类数的限制在高分辨率遥感影像中无法直接获取高精度道路网信息的问题,该文提出一种新的混合的基于支持向量机的方法:首先,利用模糊C均值聚类方法将输入的遥感影像分为3类,以减少支持向量机的错分现象;其次,运用支持向量机将不同类别的像素分为道路类和非道路类;最后,应用马尔科夫随机场对分类结果进行噪声去除,并采用形态学进行后处理,进而得到精确道路网信息。实验结果表明:该算法不仅能够从高分辨率遥感影像中提取出道路网,而且精度优于直接使用支持向量机算法以及对比算法。     

FFT-enhanced IHS transform method for fusing high-resolution satellite images

Existing image fusion techniques such as the intensity–hue–saturation (IHS) transform and principal components analysis (PCA) methods may not be optimal for fusing the new generation commercial high-resolution satellite images such as Ikonos and QuickBird. One problem is color distortion in the fused image, which causes visual changes as well as spectral differences between the original and fused images. In this paper, a fast Fourier transform (FFT)-enhanced IHS method is developed for fusing new generation high-resolution satellite images. This method combines a standard IHS transform with FFT filtering of both the panchromatic image and the intensity component of the original multispectral image. Ikonos and QuickBird data are used to assess the FFT-enhanced IHS transform method. Experimental results indicate that the FFT-enhanced IHS transform method may improve upon the standard IHS transform and the PCA methods in preserving spectral and spatial information.     

Bathymetry retrieval from optical images with spatially distributed support vector machines

This paper presents a spatially distributed support vector machine (SVM) system for estimating shallow water bathymetry from optical satellite images. Unlike the traditional global models that make predictions from a unified global model for the entire study area, our system uses locally trained SVMs and spatially weighted votes to make predictions. By using IKONOS-2 multi-spectral image and airborne bathymetric LiDAR water depth samples, we developed a spatially distributed SVM system for bathymetry estimates. The distributed model outperformed the global SVM model in predicting bathymetry from optical satellite images, and it worked well at the scenarios with a low number of training data samples. The experiments showed the localized model reduced the bathymetry estimation error by 60% from RMSE of 1.23 m to 0.48 m. Different from the traditional global model that underestimates water depth near shore and overestimates water depth offshore, the spatially distributed SVM system did not produce regional prediction bias and its prediction residual exhibited a random pattern. Our model worked well even if the sample density was much lower: The model trained with 10% of the samples was still able to obtain similar prediction accuracy as the global SVM model with the full training set.     

面向遥感目标识别耦合GA与SVM的特征优选方法

提出了GA-SVM耦合用于高分遥感目标识别的特征优选方法,将GA中的特征降维和适应度函数构建与SVM中的特征空间映射、样本训练以及分类结果在内容上耦合,利用SVM的识别结果指导GA的进化方向。同时,为减小未成熟收敛风险,对传统GA做了改进。实验表明,该方法在高分遥感影像目标识别中效果较好。     

A framework for automated change detection system

To enhance the ability of remote sensing system to provide accurate, timely, and complete geo-spatial information at regional or global scale, an automated change detection system has been and will continue to be one of the important and challenging problems in remote sensing. In this paper, the authors propose a framework for automated change detection system at landscape level using various geo-spatial data sources including multi-sensor remotely sensed imagery and ancillary data layers. In this framework, database is the central part and some associated techniques are discussed. These techniques includes five subsystems: automated feature-based image registration, automated change finding, automated change feature extraction and identification, intelligent change recognition, change accuracy assessment and database updating and visualization.     

A FRAMEWORK FOR AUTOMATED CHANGE DETECTION SYSTEM

1Ｃｕｒｒｅｎｔｃｈａｎｇｅｄｅｔｅｃｔｉｏｎｔｅｃｈ ｎｉｑｕｅｓＡｕｔｏｍａｔｉｃｃｈａｎｇｅｄｅｔｅｃｔｉｏｎｉｎｉｍａｇｅｓｏｆａｇｉｖｅｎｓｃｅｎｅａｃｑｕｉｒｅｄａｔｄｉｆｆｅｒｅｎｔｔｉｍｅｓｉｓｏｎｅｏｆｔｈｅｍｏｓｔｉｎｔｅｒｅｓｔｉｎｇｔｏｐｉｃｓｏｆｉｍａｇｅｐｒｏｃｅｓｓｉｎｇ .Ｉｔｆｉｎｄｓｉｍ ｐｏｒｔａｎｔａｐｐｌｉｃａｔｉｏｎｓｗｉｔｈｉｎｄｉｆｆｅｒｅｎｔｃｏｎｔｅｘｔｓ,ｒａｎｇ ｉｎｇｆｒｏｍｖｉｓｕａｌｓｕｒｖｅｉｌｌａｎｃｅａｎｄｖｉｄｅｏｃｏｄｉｎｇｔｏｔ…     

Automatic change detection for road networks from images based on GIS

Up to now, detailed strategies and algorithms of automatic change detection for road networks based on GIS have not been discussed. This paper discusses two different strategies of automatic change detection for images with low resolution and high resolution using old GIS data, and presents a buffer detection and tracing algorithm for detecting road from low-resolution images and a new profile tracing algorithm for detecting road from high-resolution images. For feature-level change detection (FL-CD), a so-called buffer detection algorithm is proposed to detect changes of features. Some ideas and algorithms of using GIS prior information and some context information such as substructures of road in high-resolution images to assist road detection and extraction are described in detail.     

Automatic Change Detection for Road Networks from Images Based on GIS

Up to now, detailed strategies and algorithms of automatic change detection for road networks based on GIS have not been discussed. This paper discusses two different strategies of automatic change detection for images with low resolution and high resolution using old GIS data, and presents a buffer detection and tracing algorithm for detecting road from low-resolution images and a new profile tracing algorithm for detecting road from high-resolution images. For feature-level change detection (FL-CD), a so-called buffer detection algorithm is proposed to detect changes of features. Some ideas and algorithms of using GIS prior information and some context information such as substructures of road in high-resolution images to assist road detection and extraction are described in detail.     

一种高分辨率遥感图像舰船检测方法研究

本文针对传统舰船检测方法在复杂海况条件下检测结果不佳,易造成大量虚警的问题,提出了在利用自适应分割算法进行舰船快速筛选的基础上采用方向梯度直方图(HOG)特征提取舰船的边缘结构信息,通过监督学习进行舰船检测虚警滤除的新算法。与现有算法相比,这种算法克服了由云、浪造成的舰船目标虚警滤除问题,在保证鲁棒性的同时兼顾了算法效率,并使用该方法对遥感图像进行了检测试验。     

基于OLI影像多参数设置的SVM分类研究

在遥感影像自动分类中仅使用光谱特征很难产生正确的分类,OLI影像是波段数较多的多光谱影像,如果增加纹理、几何等多种特征以提高分类精度,就会使得特征的维度很高.支持向量机善于解决小样本、非线性和高维的影像分类问题,但是核函数和参数的设置只能依靠实验来获得.文中在OLI影像中提取了23个特征,逐个测试核函数和参数值对分类结果的影响.研究的主要结论如下：RBF核的支持向量机分类精度最高,Sigmoid核支持向量机分类精度最低;核函数的选择对分类精度的影响最大;核函数和参数值的变化不会影响重要特征的使用,3种核的支持向量机分类所使用的重要特征基本一致.     

一种基于输入向量机的高光谱影像分类算法

支持向量机分类方法存在惩罚系数需要交叉验证获取、训练时间较长、支持向量个数随着训练样本数量的变化而变化,以及稳定性和稀疏性较差等问题。针对这些问题,提出了一种基于输入向量机的高光谱影像分类算法。该算法在核逻辑回归模型的基础上,采用前向贪心算法选择训练样本中的输入向量来进行模型的训练,达到稀疏的目的,提高影像的分类精度和分类效率。通过PHI和OMIS两组高光谱影像分类实验,结果表明基于输入向量机分类算法具有稳定性好、稀疏性强的优点。     

Generic rule-sets for automated detection of urban tree species from very high-resolution satellite data

The sustainable management and monitoring of urban forests is an important activity in the urbanized world, and operational approaches require information about the status of urban trees to determine the best strategy. One limitation in urban forest studies is the detection and discrimination of tree species using limited training data. Thus, this study focuses on developing generic rule sets from high-resolution WorldView-2 imagery in conjunction with spectral, spatial, colour and textural information for automated urban tree species detection. The object-based image analysis and its combination with statistical analysis of object features is utilized for this purpose. Results of attribute selection indicated that from 55 attributes, only 26 were useful to discriminate urban tree species, namely Messua ferrea L., Samanea saman and Casuarina sumatrana. Finally, the high overall accuracy, approximately 86.87% with kappa of 0.75 confirmed the transferability of the generic model.