Multi-temporal land-cover classification and change analysis with conditional probability networks: the case of Lesvos Island (Greece)

This study uses a series of Landsat images to map the main land-cover types on the Mediterranean island of Lesvos, Greece. We compare a single-year maximum likelihood classification (MLC) with a multi-temporal maximum likelihood classification (MTMLC) approach, with time-series class labels modelled using a first-order hidden Markov model comprising continuous and discrete variables. A rigorous validation scheme shows statistically significant higher accuracy figures for the multi-temporal approach. Land-cover change accuracies were also greatly improved by the proposed methodology: from 46% to 70%. The results show that when only two dates are used, the mapping of land use/cover is unreliable and a large number of the changes identified are due to the individual-year commission and omission errors.     

Comparison of time series tasseled cap wetness and the normalized difference moisture index in detecting forest disturbances

Vegetation indices and transformations have been used extensively in forest change detection studies. In this study, we processed multitemporal normalized difference moisture index (NDMI) and tasseled cap wetness (TCW) data sets and compared their statistical relationships and relative efficiencies in detecting forest disturbances associated with forest type and harvest intensity at five, two and one year Landsat acquisition intervals. The NDMI and TCW were highly correlated (>0.95 r²) for all five image dates. There was no significant difference between TCW and NDMI for detecting forest disturbance. Using either a NDMI or TCW image differencing method, when Landsat image acquisitions were 5 years apart, clear cuts could be detected with nearly equal accuracy compared to images collected 2 years apart. Partial cuts had much higher commission and omission errors compared to clear cut. Both methods had 7-8% higher commission and 12-22% higher omission error to detect hardwood disturbance when it occurred in the first year of the 2-year interval (as compared to 1-year interval). Softwood and hardwood change detection errors were slightly higher at 2-year Landsat acquisition intervals compared to 1-year interval. For images acquired 1 and 2 years apart, NDMI forest disturbance commission and omission errors were slightly lower than TCW. The NDMI can be calculated using any sensor that has near-infrared and shortwave bands and is at least as accurate as TCW for detecting forest type and intensity disturbance in biomes similar to the Maine forest, particularly when Landsat images are acquired less than 2 years apart. Where partial cutting is the most dominant harvesting system as is currently the case in northern Maine, we recommend images collected every year to minimize (particularly omission) errors. However, where clear cuts or nearly complete canopy removal occurs, Landsat intervals of up to 5 years may be nearly as accurate in detecting forest change as 1 or 2 year intervals.     

A minimum-cost thresholding technique for unsupervised change detection

We propose an automatic thresholding technique for difference images in unsupervised change detection. Such a technique takes into account the different costs that may be associated with commission and omission errors in the selection of the decision threshold. This allows the generation of maps in which the overall change-detection cost is minimized, i.e. the more critical kind of error is reduced according to end-user requirements.     

Region-driven distance regularized level set evolution for change detection in remote sensing images

Change detection is a fundamental task in the interpretation and understanding of remote sensing images. The aim is to partition the difference images acquired from multitemporal satellite images into changed and unchanged regions. Level set method is a promising way for remote sensing images change detection among the existed methods. Unfortunately, re-initialization, a necessary step in classical level set methods is known a complex and time-consuming process, which may limits their practical application in remote sensing images change detection. In this paper, we present an unsupervised change detection approach for remote sensing image based on an improved region-based active contour model without re-initialization. In order to eliminate the process for re-initialization and reduce the numerical errors caused by re-initialization, we describe an improving level set method for remote sensing images change detection. The proposed method introduced a distance regularization term into the energy function which could maintain a desired shape of the level set function and keep a signed distance profile near the zero level set. The experimental results on real multi-temporal remote sensing images demonstrate the advantages of our method in terms of human visual perception and segmentation accuracy.     

Cross-sensor change detection over a forested landscape: Options to enable continuity of medium spatial resolution measures

Landsat data have been widely used for change detection studies of forest ecosystems. Technical issues related to the longevity and quality of the Landsat-5 and -7 instruments prompted this investigation into how data from other sensors may be integrated with the existing Landsat image archive. Change maps indicating the location and extent of stand replacing disturbances occurring between 1999 and 2004 were developed using a rank-order change detection approach. The near-infrared (NIR) band from an image representing initial stand conditions (T1: Landsat-7 ETM+), and the NIR band of images acquired on subsequent dates and with different sensors (T2: ASTER, SPOT-4, and Landsat-5 TM) were selected, essentially acting as three different T2 images. Pair-wise comparisons between the T1 image and each of the T2 images required the pixel values to be sorted, ranked, and differenced; a threshold was then applied to the difference values to identify the stand replacing disturbances. The rank-order change detection approach precluded the need for an additional image normalization process. When compared to a manually interpreted map of change events, the output from the ASTER, SPOT-4, and Landsat-5 TM data were all equally effective in identifying all of the stand replacing disturbances that occurred between 1999 and the year of T2 image acquisition, and errors of commission were minimal. Important logistical limitations to cross-sensor change do exist however and include the lack of spatially or temporally extensive image archives for sensors other than Landsat, incompatible image footprints, and data cost and policy. This rank-order change detection approach is suitable for applications involving multi-temporal datasets where problems may exist due to image normalization, cross-sensor radiometric calibration, or unavailability of a desired sensor type.     

Error Analysis of Change Detection based on Land Cover in High Resolution Remote Sensing Images in Urban Areas

Land use/land cover change detection using high spatial resolution remote sensing image is an important content in land monitoring.However,the problems of shadow,image registration,threshold selection,detection method selection and image post-processing are more prominent in high-resolution images compared with that in medium and low resolution images,which result in more difficulties and uncertainties.Change detection of land cover was carried out base on aerial color images between 2009 and 2012 in Xianlin District of Nanjing,and the errors were analyzed in terms of intra-class and inter-class.The results show that the inter-class error accounted for 97.6% in the omission error,and the intra-class error accounted for 87.1% in the commission error.According to the error sources,72.6% of the false negative pixels are derived from the detection method,43.6% of the false positive pixels are come from detection method while 39.7% from radiation inconsistent.The analysis results in the paper provided reference for the development of new change detection algorithm.     

基于变化检测的多时相图像的融合算法

在分析多时相图像特点的基础上,提出了基于变化检测的多时相图像融合算法。该算法将小波变换的特征提取方法与ICA子空间映射的变化检测相结合,确定多时相图像各区域变化的强弱,根据本文提出的共生区域增长算法,以及变化检测的结果提出了多时相图像融合的方案,实现目标特征模板的提取。仿真结果表明,本文算法是比较有效的。     

A rule-based semi-automatic method to map burned areas: exploring the USGS historical Landsat archives to reconstruct recent fire history

This study presents a new semi-automatic method to map burned areas by using multi-temporal Land Remote Sensing Satellite Program (Landsat) Thematic Mapper (TM) and Enhanced TM Plus (ETM+) images. The method consists of a set of rules that are valid especially when the post-fire satellite image has been captured shortly after the fire event. The overall accuracy of the method when applied to two case studies in Mt Parnitha and Samos Island in Greece were 95.69% and 93.98%, respectively. The commission and omission errors for Mt Parnitha were 6.92% and 10.24%, while those for Samos Island were 3.97% and 8.80%, respectively. Between the two types of error, it is preferred to minimize omission errors, since commission errors can be easily identified as part of product quality assessment and algorithm tuning procedures. The rule-based approach minimizes human interventions and makes it possible to run the mapping algorithm for a series of images that would otherwise need extensive time investment. In case of failure to capture burned areas correctly, it is possible either to make some adjustments by modifying the thresholding coefficients of the rules, or to discard some of the rules, since some editing is usually required to correct errors following the automated extraction procedures. When this method was applied to a series of US Geological Survey (USGS) Landsat TM and ETM+ archived satellite images covering the periods 1984–1991 and 1999–2009, a total of 1773 fires were identified and mapped from six different scenes that covered Attica and the Peloponnese in Greece. The majority of uncaptured burned areas corresponded to fires with size classes of 0–1 ha and 1–5 ha, where the loss in capturing fire scars is generally significant. This was expected since it is possible that small fires, identified and recorded by forest authorities, may not have been captured by satellite data due to limitations arising either from the spatial resolution of the sensor or imposed by the temporal series, which do not systematically cover the full period.     

高分辨率影像配准误差对土地覆盖分类和变化检测的影响

近年来高分辨率影像技术发展迅速,土地专题信息的提取对高分辨率数据处理的质量提出了更高的要求,针对配准误差对影像处理和应用的影响研究,有助于专题信息提取过程中遥感数据处理质量控制指标的确定。选取北京市通州区不同时相的IKONOS影像作为实验数据进行模拟研究,在实验研究中通过产生具有不同配准误差的图像,从影像融合、土地覆盖分类和变化检测等角度,分析不同的配准误差对遥感应用的影响。结果表明:随着配准误差的增大,融合图像的可分辨性降低,配准误差增加到3个像元时,土地覆盖分类精度降低2~3%,土地覆盖变化检测中增加了5%的伪变化信息,虚检率增大。     

基于多时相遥感图像的人造目标变化检测算法

传统的像素级变化检测方法的检测性能受到以下因素的严重制约: 图像辐射差异、配准误差和差异图像分类门限的选取, 并且难以从检测信息中提取出关键的变化. 本文针对遥感图像中人造目标的变化检测问题, 提出了一种综合特征级和像素级的两步变化检测算法. 首先将大幅多时相遥感图像分成一系列子图像对, 采用有监督子图像对分类方法, 提取人造目标变化的感兴趣区域, 然后采用像素级变化检测算法对感兴趣区域进行变化检测, 得到定量的检测结果. 实验结果表明了该算法的可行性和有效性.     

结合MNF变换和Canny算子的遥感影像变化检测

针对大面积区域的多时相遥感影像变化检测的需求，提出了一种基于最小噪声分离（MNF）的Canny边缘检测提取影像变化信息的检测方法。对多时相影像采用多种变换组合成具有多维波段信息的影像，采用最小噪声分离法分离噪声并得到单波段差异图，通过Canny边缘检测法计算梯度幅值，采用高低双阈值法细化边缘，从而提取差异图变化边缘，有效突出了变化信息。以1995年和2003年加扎勒河的两期遥感影像为例，利用两时相影像进行土地覆被变化检测。实验结果表明，该方法适用于监测大面积区域内地物的突变情况。在数据基础上进行最小噪声分离可以有效解决传统Canny边缘检测提取边缘时造成的伪边缘现象，同时采用高低双阈值法有效去除伪边缘点，从而获得更加精确、直观的变化检测效果，在自然地理变化监测、地理国情灾害监测等有很好的应用价值。     

Accountability and automation bias

Although generally introduced to guard against human error, automated devices can fundamentally change how people approach their work, which in turn can lead to new and different kinds of error. The present study explored the extent to which errors of omission (failures to respond to system irregularities or events because automated devices fail to detect or indicate them) and commission (when people follow an automated directive despite contradictory information from other more reliable sources of information because they either fail to check or discount that information) can be reduced under conditions of social accountability. Results indicated that making participants accountable for either their overall performance or their decision accuracy led to lower rates of “automation bias”. Errors of omission proved to be the result of cognitive vigilance decrements, whereas errors of commission proved to be the result of a combination of a failure to take into account information and a belief in the superior judgement of automated aids.     

A novel approach based on structural information for change detection in SAR images

Structural information, extracted by simulating the human visual system (HVS), is independent of viewing conditions and individual observers. Structural similarity (SSIM), a measure of similarity between two images, has been widely used in image quality assessment. Given the fact that the change detection techniques identify the changed area by the similarity of multi-temporal images, SSIM has significant prospect in change detection of synthetic aperture radar (SAR) images. However, the experimental results show that SSIM performs worse in change detection of multi-temporal SAR images. In this study, we first propose an advanced SSIM (ASSIM) based on a two-step assumption of extracting structural information and a visual attention measure (VAM) model. Then, we propose a novel approach based on ASSIM for change detection in SAR images. SSIM, ASSIM, and state-of-the-art methods are tested on two datasets to compare their performances in change detection of SAR images. Experimental results show that the proposed method can acquire a better difference image than SSIM and other state-of-the-art methods, and improve the accuracy of change detection in SAR images effectively.     

基于AROP程序包的类Landsat遥感影像配准与正射纠正试验和精度分析

遥感影像的精确配准和正射纠正是进行图像融合、变化检测、图像镶嵌、定量遥感建模、多时相和多传感器影像协同应用的基础和前提。以美国国家航空和航天管理局下设LEDAPS(Landsat Ecosystem Disturbance Adaptive Processing System)课题组开发的配准与正射纠正程序包AROP(Automated Registration and Orthorectification Package)为例,详细阐述了其配准的原理与程序设计流程,并对其配准的精度进行了分析和评价。试验表明:AROP程序包算法能够找出足够的控制点,且控制点分布较为均匀,配准误差小于0.5个像元。误差特征表现为:扫描误差明显大于航向误差,误差的结果与影像漂移、DEM、坡度存在一定的相关性,高程和坡度是影响配准精度的主要因素之一。该程序包目前能够用于对我国CBERS影像的正射校正以及波段不匹配处理,但是对HJ卫星CCD影像数据配准还有待于进一步研究。     

基于引导滤波和差分图像的多聚焦图像融合方法

针对传统的多聚焦图像的空间域融合容易出现边缘模糊的问题,提出了一种基于引导滤波（GF）和差分图像的多聚焦图像融合方法。首先,将源图像进行不同水平的GF,并对滤波后图像进行差分,从而获得聚焦特征图像;随后,利用聚焦特征图像的梯度能量（EOG）信息获得初始决策图,对初始决策图进行空间一致性检查以及形态学操作以消除因EOG相近而造成的噪点;然后,对初始决策图进行GF以得到优化后决策图,从而避免融合后的图像存在边缘骤变的问题;最后,基于优化后决策图对源图像进行加权融合,以得到融合图像。选取3组经典的多聚焦图像作为实验图像,将所提方法与其他9种多聚焦图像融合方法得到的结果进行比较。主观视觉效果显示,所提方法能更好地将多聚焦图像的细节信息保存下来,另外,经该方法处理后的图像的4项客观评价指标均显著优于对比方法。结果表明,所提方法能够获得高质量的融合图像,较好地保留原始图像信息,有效解决传统多聚焦图像融合出现的边缘模糊问题。     

Analysis of spatial distribution pattern of change-detection error caused by misregistration

Misregistration between multitemporal remotely sensed images is one of the significant sources of change-detection errors. In this study, spatial distribution of change-detection errors induced by misregistration was analysed quantitatively. First, multitemporal images are registered with different misregistration values measured by root mean square error (RMSE) from 0 to 1 pixels. The image differencing method, one of the most widely used change-detection methods, is then used to detect changes. Finally, the spatial distribution pattern of change-detection errors caused by misregistration is analysed using buffering analysis based on multitemporal image edges. Experimental results indicate that the commission errors caused by misregistration values from 0 to 1 pixels are almost always within 1 pixel of the edge, regardless of image resolution. In addition, the omission errors falling within 1 pixel of the edges are about 70% for medium-resolution images. The omission errors falling within 1 or 2 pixels of the edges for high-resolution images can be as much as 50% to 60%. This work improves the understanding of spatial distribution of change-detection errors caused by misregistration and shows the relations between these errors and image edges. Moreover, it is helpful for developing new methods by combining edge and spatial information to reduce the adverse effects of misregistration on change-detection.     

Using local transition probability models in Markov random fields for forest change detection

Change detection based on the comparison of independently classified images (i.e. post-classification comparison) is well-known to be negatively affected by classification errors of individual maps. Incorporating spatial-temporal contextual information in the classification helps to reduce the classification errors, thus improving change detection results. In this paper, spatial-temporal Markov Random Fields (MRF) models were used to integrate spatial-temporal information with spectral information for multi-temporal classification in an attempt to mitigate the impacts of classification errors on change detection. One important component in spatial-temporal MRF models is the specification of transition probabilities. Traditionally, a global transition probability model is used that assumes spatial stationarity of transition probabilities across an image scene, which may be invalid if areas have varying transition probabilities. By relaxing the stationarity assumption, we developed two local transition probability models to make the transition model locally adaptive to spatially varying transition probabilities. The first model called locally adjusted global transition model adapts to the local variation by multiplying a pixel-wise probability of change with the global transition model. The second model called pixel-wise transition model was developed as a fully local model based on the estimation of the pixel-wise joint probabilities. When applied to the forest change detection in Paraguay, the two local models showed significant improvements in the accuracy of identifying the change from forest to non-forest compared with traditional models. This indicates that the local transition probability models can present temporal information more accurately in change detection algorithms based on spatial-temporal classification of multi-temporal images. The comparison between the two local transition models showed that the fully local model better captured the spatial heterogeneity of the transition probabilities and achieved more stable and consistent results over different regions of a large image scene.     

面向对象多特征融合的水域岸线目标变化检测

为认真落实河长制"清四乱"等专项行动,量化水域岸线监管测评工作,以岸线码头为目标,研究一种基于面向对象思想多特征融合的水域岸线目标变化检测方法。针对多时相高分辨率遥感影像,利用面向对象多尺度分割原理将具有空间连续性的同类区域划分为目标对象,提取目标对象的光谱、纹理及几何结构组成特征矩阵,并利用高斯径向基核函数支持向量机(RBF-SVM)进行分类;计算变化矢量差值,并与人工判读数据对比分析得到目标变化检测结果。实验结果表明,该研究应用于水域岸线上目标的变化检测中效果明显,RBFSVM分类误差影响最终目标变化检测的正确率,可为实现河湖水域岸线长效管护提供技术支撑。     

基于多时相巡检图像的变电设备抗干扰缺陷检测

近年来,变电站中广泛采用机器视觉算法分析多时相巡检图像的差异变化,用于检测各类变电设备缺陷,以确保运行安全.然而,由于拍摄时刻不同,多时相图像间存在天气、光照、季节等各类干扰变化,对变电设备的缺陷检测提出了挑战.对此,提出一种基于多时相巡检图像的变电设备抗干扰缺陷检测方法.首先,利用风格迁移模型CycleGAN学习不同风格域之间的映射关系,并基于检测图生成足量存在天气、光照、季节干扰变化的干扰图;其次,基于参考图$+$检测图$+$干扰图三元组对三重孪生网络TripleNet进行协同训练,在特征层面提出空间一致性损失以抵抗各类干扰变化,用于提取三者鲁棒的多尺度差异特征;最后,搭建特征聚合网络PANet融合多尺度差异特征,输出多尺度的缺陷检测结果.在实际变电设备多时相巡检图像数据集上进行实验验证,结果表明,所提出方法相较于非孪生网络和一般孪生网络可提升2.09%和0.67%的mAP,且在原始样本与干扰样本上的检测精度更均衡,而且所提出方法可以在提升变电设备缺陷检测模型精度的同时增强模型的抗干扰能力.     

Land-cover change detection using multi-temporal MODIS NDVI data

Monitoring the locations and distributions of land-cover changes is important for establishing links between policy decisions, regulatory actions and subsequent land-use activities. Past studies incorporating two-date change detection using Landsat data have tended to be performance limited for applications in biologically complex systems. This study explored the use of 250 m multi-temporal MODIS NDVI 16-day composite data to provide an automated change detection and alarm capability on a 1 year time-step for the Albemarle-Pamlico Estuary System (APES) region of the US. Detection accuracy was assessed for 2002 at 88%, with a reasonable balance between change commission errors (21.9%), change omission errors (27.5%), and Kappa coefficient of 0.67. Annual change detection rates across the APES over the study period (2002-2005) were estimated at 0.7% per annum and varied from 0.4% (2003) to 0.9% (2004). Regional variations were also readily apparent ranging from 1.6% to 0.1% per annum for the tidal water and mountain ecological zones, respectfully. This research included the application of an automated protocol to first filter the MODIS NDVI data to remove poor (corrupted) data values and then estimate the missing data values using a discrete Fourier transformation technique to provide high-quality uninterrupted data to support the change detection analysis. The methods and results detailed in this article apply only to non-agricultural areas. Additional limitations attributed to the coarse resolution of the NDVI data included the overestimation of change area that necessitated the application of a change area correction factor.