Monitoring Detailed Land‐Cover Changes in Shrubland Habitat Reserves Using Multi‐temporal IKONOS Data

Abstract

The objective of this study was to explore the utility of multi‐temporal, multi‐spectral image data acquired by the IKONOS satellite system for monitoring detailed land cover changes within shrubland habitat reserves. Sub‐pixel accuracy in date‐to‐date registration was achieved, in spite of the irregular relief of the study area and the high spatial resolution of the imagery. Change vector classification enabled features ranging in size from tens of square meters to several hectares to be detected and six general land cover change classes to be identified. Interpretation of the change vector classification product in conjunction with visual inspection of the multi‐temporal imagery enabled identification of specific change types such as: vegetation disturbance and associated increase in soil exposure, shrub removal, urban edge vegetation clearing and fire maintenance, increase in vegetation cover, spread of invasive plant species, fire scars and subsequent recovery, erosional scouring, trail and road development, and expansion of bicycle disturbances.     

Tree species classification using UAS-based digital aerial photogrammetry point clouds and multispectral imageries in subtropical natural forests

Tree species composition of forest stand is an important indicator of forest inventory attributes for assessing ecosystem health, understanding successional processes, and digitally displaying forest biodiversity. In this study, we acquired high spatial resolution multispectral and RGB imagery over a subtropical natural forest in southwest China using a fixed-wing UAV system. Digital aerial photogrammetric (DAP) technique was used to generate multi-spectral and RGB derived point clouds, upon which individual tree crown (ITC) delineation algorithms and a machine learning classifier were used to identify dominant tree species. To do so, the structure-from-motion method was used to generate RGB imagery-based DAP point clouds. Then, three ITC delineation algorithms (i.e., point cloud segmentation (PCS), image-based multiresolution segmentation (IMRS), and advanced multiresolution segmentation (AMRS)) were used and assessed for ITC detection. Finally, tree-level metrics (i.e., multispectral, texture and point cloud metrics) were used as metrics in the random forest classifier used to classify eight dominant tree species. Results indicated that the accuracy of the AMRS ITC segmentation was highest (F₁-score = 82.5 %), followed by the segmentation using PCS (F₁-score = 79.6 %), the IMRS exhibited the lowest accuracy (F₁-score = 78.6 %); forest types classification (coniferous and deciduous) had a higher accuracy than the classification of all eight tree species, and the combination of spectral, texture and structural metrics had the highest classification accuracy (overall accuracy = 80.20 %). In the classification of both eight tree species and two forest types, the classification accuracies were lowest when only using spectral metrics, indicated that the texture metrics and point cloud structural metrics had a positive impact on the classification (the overall accuracy and kappa accuracy increased by 1.49–4.46 % and 2.86–6.84 %, respectively).     

A Robust Texture Analysis and Classification Approach for Urban Land‐Use and Land‐Cover Feature Discrimination

Abstract

Attempts to analyze urban features and to classify land use and land cover directly from high‐resolution satellite data with traditional computer classification techniques have proven to be inefficient for two primary reasons. First, urban landscapes are composed of complex features. Second, traditional classifiers employ spectral information based on single pixel value and ignore a great amount of spatial information. Texture plays an important role in image segmentation and object recognition, as well as in interpretation of images in a variety of applications. This study analyzes urban texture features in multi‐spectral image data. Recent developments in the very powerful mathematical theory of wavelet transforms have received overwhelming attention by image analysts. An evaluation of the ability of wavelet transform in urban feature extraction and classification was performed in this study, with six types of urban land cover features classified. The preliminary results of this research indicate that the accuracy of texture analysis in classifying urban features in fine resolution image data could be significantly improved with the use of wavelet transform approach.     

融合形状和光谱的高空间分辨率遥感影像分类

提出了一种像元形状指数及基于形状和光谱特征融合的高（空间）分辨率遥感影像分类方法。形状和光谱是遥感影像纹理的具体表现形式,尤其在高分辨率影像中地物细节得到充分表达,相邻像元的关系及其共同表征的形状特性成为分类的重要因素。本文用像元及其邻域的关系来描述其空间结构,同时为了更全面地利用影像特征,提出了基于支持向量机的形状和光谱融合分类方法。实验证明,该方法计算简便且能有效表达高分辨率影像的地物特征,提高分类精度。     

Comparison of four UAV georeferencing methods for environmental monitoring purposes focusing on the combined use with airborne and satellite remote sensing platforms

This work is aimed at the environmental remote sensing community that uses UAV optical frame imagery in combination with airborne and satellite data. Taking into account the economic costs involved and the time investment, we evaluated the fit-for-purpose accuracy of four positioning methods of UAV-acquired imagery: 1) direct georeferencing using the onboard raw GNSS (GNSSNAV) data, 2) direct georeferencing using Post-Processed Kinematic single-frequency carrier-phase without in situ ground support (PPK1), 3) direct georeferencing using Post-Processed Kinematic double-frequency carrier-phase GNSS data with in situ ground support (PPK2), and 4) indirect georeferencing using Ground Control Points (GCP). We tested a multispectral sensor and an RGB sensor, onboard multicopter platforms. Orthophotomosaics at <0.05 m spatial resolution were generated with photogrammetric software. The UAV image absolute accuracy was evaluated according to the ASPRS standards, wherein we used a set of GCPs as reference coordinates, which we surveyed with a differential GNSS static receiver. The raw onboard GNSSNAV solution yielded horizontal (radial) accuracies of RMSEr≤1.062 m and vertical accuracies of RMSEz≤4.209 m; PPK1 solution gave decimetric accuracies of RMSEr≤0.256 m and RMSEz≤0.238 m; PPK2 solution, gave centimetric accuracies of RMSEr≤0.036 m and RMSEz≤0.036 m. These results were further improved by using the GCP solution, which yielded accuracies of RMSEr≤0.023 m and RMSEz≤0.030 m. GNSSNAV solution is a fast and low-cost option that is useful for UAV imagery in combination with remote sensing products, such as Sentinel-2 satellite data. PPK1, which can register UAV imagery with remote sensing products up to 0.25 m pixel size, as WorldView-like satellite imagery, airborne lidar or orthoimagery, has a higher economic cost than the GNSSNAV solution. PPK2 is an acceptable option for registering remote sensing products of up to 0.05 m pixel size, as with other UAV images. Moreover, PPK2 can obtain accuracies that are approximate to the usual UAV pixel size (e.g. co-register in multitemporal studies), but it is more expensive than PPK1. Although indirect georeferencing can obtain the highest accuracy, it is nevertheless a time-consuming task, particularly if many GCPs have to be placed. The paper also provides the approximate cost of each solution.     

一种基于概率潜在语义模型的高分辨率遥感影像分类方法

针对高分辨率遥感影像中"同谱异物","同物异谱"现象对影像分类过程造成的干扰,将文本分析中的概率潜在语义模型应用于高分辨率遥感影像分类,提出一种无监督的遥感影像分类新方法.该方法首先利用均值漂移分割方法对影像进行分割构建图像区域集合,然后提取集合各区域中每个像元的Gabor纹理特征,并对这些特征进行聚类形成视觉词汇,最...     

Markov-random-field-based super-resolution mapping for identification of urban trees in VHR images

Identification of tree crowns from remote sensing requires detailed spectral information and submeter spatial resolution imagery. Traditional pixel-based classification techniques do not fully exploit the spatial and spectral characteristics of remote sensing datasets. We propose a contextual and probabilistic method for detection of tree crowns in urban areas using a Markov random field based super resolution mapping (SRM) approach in very high resolution images. Our method defines an objective energy function in terms of the conditional probabilities of panchromatic and multispectral images and it locally optimizes the labeling of tree crown pixels. Energy and model parameter values are estimated from multiple implementations of SRM in tuning areas and the method is applied in QuickBird images to produce a 0.6 m tree crown map in a city of The Netherlands. The SRM output shows an identification rate of 66% and commission and omission errors in small trees and shrub areas. The method outperforms tree crown identification results obtained with maximum likelihood, support vector machines and SRM at nominal resolution (2.4 m) approaches.     

Multi-Feature Classification Approach for High Spatial Resolution Hyperspectral Images

High spatial resolution hyperspectral images not only contain abundant radiant and spectral information, but also display rich spatial information. In this paper, we propose a multi-feature high spatial resolution hyperspectral image classification approach based on the combination of spectral information and spatial information. Three features are derived from the original high spatial resolution hyperspectral image: the spectral features that are acquired from the auto subspace partition technique and the band index technique; the texture features that are obtained from GLCM analysis of the first principal component after principal component analysis is performed on the original image; and the spatial autocorrelation features that contain spatial band X and spatial band Y, with the grey level of spatial band X changing along columns and the grey level of spatial band Y changing along rows. The three features are subsequently combined together in Support Vector Machine to classify the high spatial resolution hyperspectral image. The experiments with a high spatial resolution hyperspectral image prove that the proposed multi-feature classification approach significantly increases classification accuracies.     

Implications of sensor design for coral reef detection: Upscaling ground hyperspectral imagery in spatial and spectral scales

Remote sensing offers a potential tool for large scale environmental surveying and monitoring. However, remote observations of coral reefs are difficult especially due to the spatial and spectral complexity of the target compared to sensor specifications as well as the environmental implications of the water medium above. The development of sensors is driven by technological advances and the desired products. Currently, spaceborne systems are technologically limited to a choice between high spectral resolution and high spatial resolution, but not both. The current study explores the dilemma of whether future sensor design for marine monitoring should prioritise on improving their spatial or spectral resolution. To address this question, a spatially and spectrally resampled ground-level hyperspectral image was used to test two classification elements: (1) how the tradeoff between spatial and spectral resolutions affects classification; and (2) how a noise reduction by majority filter might improve classification accuracy. The studied reef, in the Gulf of Aqaba (Eilat), Israel, is heterogeneous and complex so the local substrate patches are generally finer than currently available imagery. Therefore, the tested spatial resolution was broadly divided into four scale categories from five millimeters to one meter. Spectral resolution resampling aimed to mimic currently available and forthcoming spaceborne sensors such as (1) Environmental Mapping and Analysis Program (EnMAP) that is characterized by 25 bands of 6.5 nm width; (2) VENμS with 12 narrow bands; and (3) the WorldView series with broadband multispectral resolution. Results suggest that spatial resolution should generally be prioritized for coral reef classification because the finer spatial scale tested (pixel size < 0.1 m) may compensate for some low spectral resolution drawbacks. In this regard, it is shown that the post-classification majority filtering substantially improves the accuracy of all pixel sizes up to the point where the kernel size reaches the average unit size (pixel < 0.25 m). However, careful investigation as to the effect of band distribution and choice could improve the sensor suitability for the marine environment task. This in mind, while the focus in this study was on the technologically limited spaceborne design, aerial sensors may presently provide an opportunity to implement the suggested setup.     

Texture classification of Mediterranean land cover

Maximum likelihood (ML) and artificial neural network (ANN) classifiers were applied to three Landsat Thematic Mapper (TM) image sub-scenes (termed urban, agricultural and semi-natural) of Cukurova, Turkey. Inputs to the classifications comprised (i) spectral data and (ii) spectral data in combination with texture measures derived on a per-pixel basis. The texture measures used were: the standard deviation and variance and statistics derived from the co-occurrence matrix and the variogram. The addition of texture measures increased classification accuracy for the urban sub-scene but decreased classification accuracy for agricultural and semi-natural sub-scenes. Classification accuracy was dependent on the nature of the spatial variation in the image sub-scene and, in particular, the relation between the frequency of spatial variation and the spatial resolution of the imagery. For Mediterranean land, texture classification applied to Landsat TM imagery may be appropriate for the classification of urban areas only.     

Land use change detection based on multi‐date imagery from different satellite sensor systems

An empirical study was performed assessing the accuracy of land use change detection when using satellite image data acquired ten years apart by sensors with differing spatial resolutions. Landsat/Multi‐spectral Scanner (MSS) with Landsat/Thematic Mapper (TM) or SPOT/High Resolution Visible (HRV) multi‐spectral (XS) data were used as a multi‐data pair for detecting land use change. The primary objectives of the study were to: (1) compare standard change detection methods (e.g. multi‐date ratioing and principal components analysis) applied to image data of varying spatial resolution; (2) assess whether to transform the raster grid of the higher resolution image data to that of the lower resolution raster grid or vice‐versa in the registration process: and (3) determine if Landsat/TM or SPOT/ HRV(XS) data provides more accurate detection of land use changes when registered to historical Landsat/MSS data.

Ratioing multi‐sensor, multi‐date satellite image data produced higher change detection accuracies than did principal components analysis and is useful as a land use change enhancement technique. Ratioing red and near infrared bands of a Landsat/MSS‐SPOT/HRV(XS) multi‐date pair produced substantially higher change detection accuracies (～10%) than ratioing similar bands of a Landsat/MSS ‐ Landsat/TM multi‐data pair. Using a higher‐resolution raster grid of 20 meters when registering Landsat/MSS and SPOTZHRV(XS) images produced a slightly higher change detection accuracy than when both images were registered to an 80 meter raster grid. Applying a “majority”; moving window filter whose size approximated a minimum mapping unit of 1 hectare increased change detection accuracies by 1–3% and reduced commission errors by 10–25%.     

Identification of hazelnut fields using spectral and Gabor textural features

Land cover identification and monitoring agricultural resources using remote sensing imagery are of great significance for agricultural management and subsidies. Particularly, permanent crops are important in terms of economy (mainly rural development) and environmental protection. Permanent crops (including nut orchards) are extracted with very high resolution remote sensing imagery using visual interpretation or automated systems based on mainly textural features which reflect the regular plantation pattern of their orchards, since the spectral values of the nut orchards are usually close to the spectral values of other woody vegetation due to various reasons such as spectral mixing, slope, and shade. However, when the nut orchards are planted irregularly and densely at fields with high slope, textural delineation of these orchards from other woody vegetation becomes less relevant, posing a challenge for accurate automatic detection of these orchards. This study aims to overcome this challenge using a classification system based on multi-scale textural features together with spectral values. For this purpose, Black Sea region of Turkey, the region with the biggest hazelnut production in the world and the region which suffers most from this issue, is selected and two Quickbird archive images (June 2005 and September 2008) of the region are acquired. To differentiate hazel orchards from other woodlands, in addition to the pansharpened multispectral (4-band) bands of 2005 and 2008 imagery, multi-scale Gabor features are calculated from the panchromatic band of 2008 imagery at four scales and six orientations. One supervised classification method (maximum likelihood classifier, MLC) and one unsupervised method (self-organizing map, SOM) are used for classification based on spectral values, Gabor features and their combination. Both MLC and SOM achieve the highest performance (overall classification accuracies of 95% and 92%, and Kappa values of 0.93 and 0.88, respectively) when multi temporal spectral values and Gabor features are merged. High F_β values (a combined measure of producer and user accuracy) for detection of hazel orchards (0.97 for MLC and 0.94 for SOM) indicate the high quality of the classification results. When the classification is based on multi spectral values of 2008 imagery and Gabor features, similar F_β values (0.95 for MLC and 0.93 for SOM) are obtained, favoring the use of one imagery for cost/benefit efficiency. One main outcome is that despite its unsupervised nature, SOM achieves a classification performance very close to the performance of MLC, for detection of hazel orchards.     

基于SVM的高光谱遥感图像亚像元定位

提出了基于支持向量机（support vector machine,SVM）的高光谱遥感图像亚像元定位方法。全变分（total variation,TV）模型是经典的保边缘平滑滤波器,本文将其引入作为预处理,来提高混合像元分解及亚像元定位的精度;本文方法在训练和检验样本的构建过程中,依据空间相关性理论,同时考虑了中心像元及其邻近像元丰度值对亚像元类别归属的影响;在监督分类训练和检验过程中,通过剔除纯净像元来缩减样本数量,在保证算法准确性的同时提高了效率。对真实高光谱遥感数据进行了实验,主观评价和定量分析验证了本文方法的有效性。     

面向对象的多尺度加权联合稀疏表示的高空间分辨率遥感影像分类

针对高空间分辨率遥感影像中的地物具有多尺度特性,以及各个尺度的对象特征对地物分类精度的影响具有较强的尺度效性,并结合面向对象影像分析方法和多尺度联合稀疏表示方法在高空间分辨率遥感影像分类中的各自优点,提出了一种面向对象的多尺度加权稀疏表示的高空间分辨率遥感影像分类算法。首先,采用多尺度分割算法获得多尺度分割结果并提取对象的多尺度特征;然后,根据影像对象的多尺度分割质量测度计算各尺度的对象权重,构建面向对象的多尺度加权联合稀疏表示模型;最后,采用2个国产GF-2高空间分辨率遥感数据集和1个高光谱-高空间分辨率航空遥感数据集（WashingtonD.C.数据）验证该算法的有效性。试验结果表明,与SVM、像素级稀疏表示、单尺度和多尺度对象级稀疏表示和深度学习等算法相比较,本文算法获得了较高的OA和Kappa分类精度,提高了各个尺度地物的分类精度,有效抑止了地物分类结果中的椒盐噪声现象,同时保持大尺度地物的区域性和小尺度地物的细节信息。     

基于MERIS影像空间、光谱及时间维信息提取亚像元级的土地覆盖组分

及时获取有效的土地覆盖信息是地球系统模拟的基础。因此,中等空间分辨率传感器如MODIS或MERIS空前的通道设置与观测能力,使其具有快速更新土地覆盖图的能力。本文说明了如何结合MERIS的空间维（像元大小为300m）、光谱维（可见光与近红外范围内15个通道）和时间维（重返周期2—3d）,用于获取不同区域土地覆被组分的亚像元级组成权重。利用4月、7月和8月三期MERIS FR1b级数据得到荷兰主要土地覆被类型的组成权重。单一时相和多时相的数据都使用单个像元最优化的端元数进行线性光谱分解。利用一种形态偏离指数得到MERIS的空间维并用于端元的选择。应用荷兰土地利用数据库（LGN5）25m分辨率的栅格数据作为本文的参考数据。基于这种数据的高分辨率,因此可以从像元和亚像元的水平同时评价的分类精度。结果显示,结合4月和7月的影像可以获得最优的分类结果,精度约为58%。总的说来,亚像元和像元级的分类精度相似。通过几种组分类别和日期的光谱融合表明,物候状况对于数据获取时相最佳结合的选择以及正确识别土地覆盖类型的重要性。     

Impact of the spatial resolution on the energy balance components on an open-canopy olive orchard

The recent technical improvements in the sensors used to acquire images from land surfaces has made possible to assess the performance of the energy balance models using unprecedented spatial resolutions. Thus, the objective of this work is to evaluate the response of the different energy balance components obtained from METRIC model as a function of the input pixel size. Very high spatial resolution airborne images (≈50 cm) on three dates over olive orchards were used to aggregate different spatial resolutions, ranging from 5 m to 1 km. This study represents the first time that METRIC model has been run with such high spatial resolution imagery in heterogeneous agricultural systems, evaluating the effects caused by its aggregation into coarser pixel sizes. Net radiation and soil heat flux showed a near insensitive behavior to spatial resolution changes, reflecting that the emissivity and albedo respond linearly to pixel aggregation. However, greater discrepancies were obtained for sensible (up to 17%) and latent (up to 23%) heat fluxes at spatial resolutions coarser than 30 × 30 m due to the aggregation of non-linear components, and to the inclusion of non-agricultural areas in such aggregation. Results obtained confirm the good performance of METRIC model when used with high spatial resolution imagery, whereas they warn of some major errors in crop evapotranspiration estimation when medium or large scales are used.     

Land Cover/Land use Mapping Using Different Wavelet Packet Transforms for LISS IV Madurai Imagery

Texture in high resolution satellite images requires substantial improvement in the conventional segmentation algorithms. The use of wavelet packet transforms for texture analysis and image classification of high spatial resolution LISS IV imagery provide more details about the urban areas. This paper analyses the performance of a combination of Wavelet Packet Statistical Features (WPSFs) and Wavelet Packet Co-occurrence Features (WPCFs) for the classification of LISS IV images. The classification accuracy per pixel is improved in this paper by varying the window size. Four indices—user’s accuracy, producer’s accuracy, overall accuracy and kappa co-efficient are used to assess the accuracy of the classified data. Experimental results show that a multi-band and multi-level wavelet packet approach can be used to drastically increase the classification accuracy.     

基于独立分量分析的遥感影像非监督分类方法

利用独立分量分析的方法，从图像信号分离的角度出发，将每个波段像元的光谱特征看成是由相互独立的不同地物类型光谱信号混合而成。通过ETM^-遥感影像数据的分类试验，验证了该方法应用于多光谱遥感影像非监督分类的有效性。     

基于独立分量分析的纹理特征维数减少

提出了基于ICA纹理特征维数减少的方法，通过QuickBird多光谱影像的实验证明，ICA对各种纹理特征降维的普适性最强，类别可分性最高。     

Segmentation and classification of high spatial resolution images based on Hölder exponents and variance

Pixel-based or texture-based classification technique individually does not yield an appropriate result in classifying the high spatial resolution remote sensing imagery since it comprises textured and non-textured regions. In this study, Hölder exponents (HE) and variance (VAR) are used together to transform the image for measuring texture. A threshold is derived to segment the transformed image into textured and non-textured regions. Subsequently, the original image is extracted into textured and non-textured regions using this segmented image mask. Afterward, extracted textured region is classified using ISODATA classification algorithm considering HE, VAR, and intensity values of individual pixel of textured region. And extracted non-textured region of the image is classified using ISODATA classification algorithm. In case of non-textured region, HE and VAR value of individual pixel is not considered for classification for significant textural variation is not found among different classes. Consequently, the classified outputs of non-textured and textured regions that are generated independently are merged together to get the final classified image. IKONOS 1 m PAN images are classified using the proposed algorithm, and the classification accuracy is more than 88%.