Optimizing the spatial resolution of WorldView-2 imagery for discriminating forest vegetation at subspecies level in KwaZulu-Natal,South Africa

The objective of this study was to identify an appropriate spatial resolution for discriminating forest vegetation at subspecies level. WorldView-2 imagery was progressively resampled to coarser spatial resolutions. At a compartment level, 30 × 30-m subsets were generated across forest compartments to represent the five forest subspecies investigated in this study. From the centre of each subset, the spatial resolution of the original WorldView-2 image was resampled from 6 to 34-m, with increments of 4-m. The variance was then calculated at every resampled spatial resolution using each of the eight WorldView-2 bands. Based on the sampling theorem, the 3-m spatial resolution provided an appropriate resolution for all subspecies investigated. The WorldView-2 image was subsequently classified using the partial least squares linear discriminant analysis algorithm and the appropriate spatial resolution. An overall classification accuracy of 90% was established with an allocation disagreement of 9 and a quantity disagreement of 1.     

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%.     

遥感尺度内涵及遥感像元尺度效应分析的改进分形方法

本文首先分析了遥感影像尺度的三层次内涵。重点针对遥感像元尺度,分析了遥感像元尺度效应及其分形机理,由于现有分形方法没有考虑影像本身尺度(空间分辨率),造成尺度间分形维数的比较时像元尺度效应变化难以有效反映,本文针对此问题提出了基于表面积的加窗分形布朗运动和加窗双层地毯两种改进分形方法。为验证改进分形方法的可靠性,采用了不同像元尺度下系列监督分类进行验证。试验结果表明,每种地物的分维数都随着空间分辨率的降低或像元尺度的缩小,呈总体下降趋势,在某些特征尺度上会出现预示着某些地物结构的拐点,这些拐点对观测该区域地物具有一定指示意义。系列监督分类精度也一定程度上证明了以上两种改进分形方法在分析尺度效应中的可行性。因此本文的方法对于分析遥感像元尺度效应和探索地物尺度聚合规律具有一定的理论意义。     

Using high resolution QuickBird imagery for crop identification and area estimation

Imagery from recently launched high spatial resolution satellite sensors offers new opportunities for crop assessment and monitoring. A 2.8-m multispectral QuickBird image covering an intensively cropped area in south Texas was evaluated for crop identification and area estimation. Three reduced-resolution images with pixel sizes of 11.2 m, 19.6 m, and 30.8 m were also generated from the original image to simulate coarser resolution imagery from other satellite systems. Supervised classification techniques were used to classify the original image and the three aggregated images into five crop classes (grain sorghum, cotton, citrus, sugarcane, and melons) and five non-crop cover types (mixed herbaceous species, mixed brush, water bodies, wet areas, and dry soil/roads). The five non-crop classes in the 10-category classification maps were then merged as one class. The classification maps were filtered to remove the small inclusions of other classes within the dominant class. For accuracy assessment of the classification maps, crop fields were ground verified and field boundaries were digitized from the original image to determine reference field areas for the five crops. Overall accuracy for the unfiltered 2.8-m, 11.2-m, 19.6-m, and 30.8-m classification maps were 71.4, 76.9, 77.1, and 78.0%, respectively, while overall accuracy for the respective filtered classification maps were 83.6, 82.3, 79.8, and 78.5%. Although increase in pixel size improved overall accuracy for the unfiltered classification maps, the filtered 2.8-m classification map provided the best overall accuracy. Percentage area estimates based on the filtered 2.8-m classification map (34.3, 16.4, 2.3, 2.2, 8.0, and 36.8% for grain sorghum, cotton, citrus, sugarcane, melons, and non-crop, respectively) agreed well with estimates from the digitized polygon map (35.0, 17.9, 2.4, 2.1, 8.0, and 34.6% for the respective categories). These results indicate that QuickBird imagery can be a useful data source for identifying crop types and estimating crop areas.     

Satellite Data Classification Using Open Source Support

In this study we explored the potential of open source data mining software support to classify freely available Landsat image. The study identified several major classes that can be distinguished using Landsat data of 30 m spatial resolution. Decision tree classification (DTC) using Waikato environment for knowledge analysis (WEKA), open source software is used to prepare land use land cover (LULC) map and the result is compared with supervised (maximum likelihood classifier – MLC) and unsupervised (Iterative self-organizing data analysis technique - ISODATA clustering) classification techniques. The accuracy assessment indicates highest accuracy of the map prepared using DTC with overall accuracy (OA) 92 % (kappa = 0.90) followed by MLC with OA 88 % (kappa = 0.84) and ISODATA OA 76 % (kappa = 0.69). Results indicate that data set with a good definition of training sites can produce LULC map having good overall accuracy using decision tree. The paper demonstrates utility of open source system for information extraction and importance of DTC algorithm.     

Evaluation of the spatial changes in seagrass cover in the lagoons of Lakshadweep islands,India, using IRS LISS III satellite images

A study was conducted in Lakshadweep islands to determine the feasibility of using Indian Remote Sensing (IRS) satellites for detecting changes in the seagrass from other coastal features. IRS ID and IRS P6 LISS III having spatial resolution of 23.5 m with lower cost compared to all other contemporary satellites with the same spatial resolution have not been widely used for monitoring the changes in seagrass cover. In this context, the present study attempted to explore the effectiveness of LISS III data for mapping seagrasses and to inform the international community about the usefulness of these low-cost imageries for coastal resource monitoring. Supervised classification and change detection studies found a significant decrease in seagrass cover of 73.03 ha in the Lakshadweep group of islands. An overall accuracy of 67.5% was obtained for the change maps, and seagrass cover and its changes vary at different islands.     

Mapping spiny aster infestations with QuickBird imagery

QuickBird satellite imagery acquired in June 2003 and September 2004 was evaluated for detecting the noxious weed spiny aster [Leucosyris spinosa (Benth.) Greene] on a south Texas, USA rangeland area. A subset of each of the satellite images representing a diversity of cover types was extracted and used as a study site. The satellite imagery had a spatial resolution of 2.8 m and contained 11-bit data. Unsupervised and supervised classification techniques were used to classify false colour composite (green, red, and near-infrared bands) images of the study site. Imagery acquired in June was superior to that obtained in September for distinguishing spiny aster infestations. This was attributed to differences in spiny aster phenology between the two dates. An unsupervised classification of the June image showed that spiny aster had producer's and user's accuracies of 90% and 93.1%, respectively, whereas a supervised classification of the June image had producer's and user's accuracies of 90% and 81.8%, respectively. These results indicate that high resolution satellite imagery coupled with image analysis techniques can be used successfully for detecting spiny aster infestations on rangelands.     

天宫一号高光谱遥感场景分类数据集及应用

天宫一号高光谱成像仪具有空间分辨率高、光谱分辨率高、图谱合一等特性，在中国航天高光谱领域具有里程碑的意义。针对一般遥感场景分类数据集尺度单一、光谱分辨率较低等问题，本文提出基于天宫一号的多谱段、高空间分辨率、多时相高光谱遥感场景分类数据集(TG1HRSSC)。利用天宫一号高光谱成像仪获取的高质量数据，经过辐射校正、几何校正、空间裁剪、波段筛选、数据质量分析与控制等，制作了一批通用的航天高光谱遥感场景分类数据集，通过载人航天空间应用数据推广服务平台（http://www.msadc.cn[2019-09-10]）进行分发和共享。该数据集包括天宫一号高光谱成像仪获取的城镇、农田、林地、养殖塘、荒漠、湖泊、河流、港口、机场等9个典型地物场景的204个高光谱影像数据，其中5 m分辨率全色谱段1个波段、10 m分辨率可见近红外谱段54个有效波段以及20 m分辨率短波红外谱段52个有效波段。研究利用AlexNet、VGG-VD-16、GoogLeNet等深度学习算法网络对构建的数据集进行场景分类的试验，结果表明该数据集的场景分类应用实现较好效果。由于该数据集具备高分辨、高光谱等特征优势，未来在语义理解、多目标检测等方面有着广泛的应用价值。     

Spatial analysis of high-resolution urban thermal patterns in Vojvodina,Serbia

Main objective of this study was to establish a relationship between land cover and land surface temperature (LST) in urban and rural areas. The research was conducted using Landsat, WorldView-2 (WV-2) and Digital Mapping Camera. Normalised difference vegetation index and normalised difference built-up index were used for establishing the relation between built-up area, vegetation cover and LST for spatial resolution of 30 m. Impervious surface and vegetation area generated from Digital Mapping Camera from Intergraph and WV-2 were used to establish the relation between built-up area, vegetation cover and LST for spatial resolutions of 0.1, 0.5 and 30 m. Linear regression models were used to determine the relationship between LST and indicators. Main contribution of this research is to establish the use of combining remote sensing sensors with different spectral and spatial resolution for two typical settlements in Vojvodina. Correlation coefficients between LST and LST indicators ranged from 0.602 to 0.768.     

Rule-based classification of high-resolution imagery over urban areas in New York City

This paper describes the integration of results from different feature extraction algorithms using spectral and spatial attributes to detect specific urban features. Methodology includes segmentation of IKONOS data, computing attributes for creating image objects and classifying the objects with fuzzy logic and rule-based algorithms. Previous research reported low class accuracies for two specific classes – dark and grey roofs. A modified per-field approach was employed to extract urban features. New rule-sets were used on image objects having similar or near-similar spectral and spatial characteristics. Different algorithms using spectral and spatial attributes were developed to extract specific urban features from a time-series of Multi-Spectral Scanner (MSS) (4 m × 4 m) IKONOS data. The modified approach resulted in a remarkable improvement in the accuracy of classes that registered low spectral seperability and therefore low accuracy. The spectral and spatial based classification model may be useful in mapping heterogeneous and spectrally similar urban features.     

Mapping the Extent of Agroforestry Area in Different Altitudes of Tehri District,North Western Himalaya,India Through GIS and Remote Sensing Data

Assessment of area under agroforestry in Tehri district of North Western Himalaya, Uttarakhand, India has been done using GIS and remote sensing technology. The study district characterized by hilly terrain with varying elevations from 288 m to more than 2800 m and generally gentle slopes, valleys, flat land covers and agricultural terraces. High-resolution satellite imageries (spatial resolution 5.8 m) were used in this study for land uses and land covers classification. According to unsupervised classification, highest area was found under forest class (65.22%) followed by cropland (20.41%). Considerable area was also found under snow cover (9.45%) in the district. Area under agroforestry was estimated to be 5572.26 ha (1.53%) by this method, whereas it was estimated to be 7029.06 ha (1.93%) by supervised classification. Estimated cropland area comes out to be about 20.0%. An accuracy of 86.5% was found in this classification for agroforestry class. Highest area under agroforestry of 3707.36 ha was obtained in 1200–2000 m elevations followed by 2231.26 ha in 288–1200 m elevations. Negligible area was found on high elevation zones of more than 2800 m. The major agroforestry systems of dominated by Grewia oppositifolia (Bhimal), Celtis australis (Kharik) and Quercus leucotrichophora (Banj) were identified and mapped and remaining systems were grouped as others class. Estimated area under G. oppositifolia, C. australis and Q. leucotrichophora based systems come out to be 2330.82, 1456.80 and 1129.10 ha, respectively. These systems are multiple usufructs are food, fuelwood, fodder, fiber and small timber. It has been observed from the accuracy assessment that the estimates of area under agroforestry obtained under this study are reliable.     

Uncertainty analysis of soil erosion modelling using different resolution of open-source DEMs

The Digital Elevation Model (DEM) is one of the important parameters of soil erosion assessment and notable uncertainties are found in using different resolutions of the DEM. Revised Universal Soil Loss Equation model has been applied to analyze the effect of open-source DEMs with different resolution and accuracy on the uncertainties of soil erosion modelling in a part of the Narmada river basin in Madhya Pradesh in central India. Selected open-source DEMs are GTOPO30 (1 km), SRTM (30 and 90 m), CARTOSAT (30 m) and ASTER (30 m), used for estimating erosion rate. Results with better accuracy are achieved with the high-resolution DEMs (30 m) with higher vertical accuracy than the coarse resolution DEMs with lower accuracy. This study has presented potential uncertainties introduced by the open-source DEMs in soil erosion modelling for better understanding of appropriate selection and acceptable errors for researchers.     

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.     

Reclamative grouping of ravines using Cartosat-1 PAN stereo data

Information on the depth and bed width of ravines (network of gullies) at large scales is critical for their reclamation and management. Hitherto such information has been generated from aerial photographs and space borne stereo images with medium to coarse ‘z’ – axis resolution. The present study, aims at demonstrating the potential of Cartosat ?1 (an Indian Earth observations satellite) stereo images with 2.5 m spatial resolution in deriving morphometric information on ravines for their reclamative grouping. The study area is a part of Jhansi and Hamirpur districts of Uttar Pradesh, northern India. The approach involves acquiring precise ground control points using Differential GPS (DGPS), triangulation, DEM extraction and generation of ortho image as well as anaglyphs for stereo viewing. The depth and bed width of ravines were measured in the field for validation. A comparison with field observations reveal that the bed width of ravines and depth can be measured successfully with Carto-1 stereo data. The anaglyph data was used to delineate various categories of ravines based on their depth and bed width. Results indicate that the Cartosat-1 stereo images are quite suitable for delineation of three categories of ravines, namely shallow (<3 m deep and <18 m bed width), medium deep (3–9 m deep and >18 m bed width) and deep (>9 m deep) which are important for their reclamation.     

Automated analysis of ultra high resolution remote sensing data for biotope type mapping: new possibilities and challenges

The advent of very high-resolution satellite programs and digital airborne cameras with ultra high resolution offers new possibilities for very accurate mapping of the environment. With these sensors of improved spatial resolution, however, the user community faces a new problem in the analysis of this type of image data. Standard classification techniques have to be augmented with appropriate analysis procedures because the required homogeneity of landuse/landcover classes can no longer be achieved by the integration effect of large pixel sizes (e.g., 20–80 m). New intelligent techniques will have to be developed that make use of multisensor approaches, geographic information system (GIS) integration and context-based interpretation schemes.The ideal goal should be that GIS ‘intelligence’ (e.g., object and analysis models) should be used to automate the classification process. In return, GIS objects can be extracted from a remote sensing image to update the GIS database. This paper presents the development of an automated procedure for biotope type mapping from ultra high-resolution airborne scanner data (HRSC-A). The hierarchical procedure incorporates a priori GIS information, a digital surface model (DSM) and multispectral image data. The results of this study will serve as a basis for a continuous environmental monitoring process in the tidally influenced region of the Elbe River, Germany.     

结合L0平滑和超像素的高空间遥感影像非监督分类

针对基于像元的非监督分类方法对高空间遥感影像分类时易形成“椒盐”噪声和产生大量错分、漏分的问题,提出了一种结合L0平滑和超像素的非监督分类方法．首先采用L0算法对高空间遥感影像进行平滑操作,减少大量图像噪声及冗余信息;然后采用简单的线性迭代聚类（SLIC）超像素方法处理平滑后图像,进一步抑制椒盐现象的同时降低处理复杂度,得到初始聚类图;最后采用K-means非监督分类方法得到最终分类结果图．为验证本文提出的方法,选取3景高空间遥感影像作为实验数据．试验结果表明,采用提出的方法能准确对地物分类,且总体精度分别达到了72.46％、77.55％和78.44％,Kappa系数分别达到0.788、0.779和0.779．提出方法能有效解决分类中存在的“椒盐”现象,可提高分类精度,对高空间遥感影像分类具有一定的参考价值．      

Towards automated 10–30 m resolution land cover mapping in insular South-East Asia

In this study, we create and critically analyse an automated decision tree classification approach for regional level land cover mapping in insular South-East Asian conditions, using a combination of 10–30 m resolution optical and radar data. The resulting map contains 11 land cover classes and reveals a great deal of contextual information due to high spatial resolution. A limited accuracy assessment indicates 59–97% class wise accuracies. The unprecedented spatial detail of closed canopy oil palm mapping (with user’s accuracy of 90%) is seen as the most promising feature of the mapping approach. The incapability of separating primary forests from other tree cover, and the large variety of different landscapes (e.g. home gardens and tea plantations) classified as shrubland, are considered the main areas for future improvement. Overall, the study demonstrates the great potential of multi-source 10–30 m resolution high data volume land cover mapping approaches in insular South-East Asian conditions.     

Estimation of Above Ground Biomass Using Texture Metrics Derived from IRS Cartosat-1 Panchromatic Data in Evergreen Forests of Western Ghats,India

Assessment of above ground forest biomass (AGB) is essential in carbon modelling studies to provide mitigation strategies as demonstrated by reducing emissions from deforestation and forest degradation. Several researchers have demonstrated the use of remote sensing data in spatial AGB estimation, in terms of spectral and radar backscatter based approaches at a landscape scale with several known limitations. However, these methods lacked the predictive ability at high biomass ranges due to saturation. The current study addresses the problem of saturation at high biomass ranges using canopy textural metric from high resolution optical data. Fourier transform based textural ordination (FOTO) technique, which involves deriving radial spectrum information via 2D fast Fourier transform and ordination through principal component analysis was used for characterizing the textural properties of forest canopies. In the current study, plot level estimated AGB from 15 (1 ha) plots was used to relate with texture derived information from very high resolution datasets (viz., IKONOS and Cartosat-1). In addition to the estimation of high biomass ranges, one of the prime objective of the current study is to understand the effects of spatial resolution on deriving textural-AGB relationship from 2.5 m IRS Cartosat data (Cartosat-A, viewing angle = ?5°) to that of IKONOS imagery with near nadir view. Further, since texture is impacted by several illumination geometry issues, the effect of viewing geometry on the relationship was evaluated using Cartosat-F (Viewing angle = 26°) imagery. The results show that the FOTO method using stereo Cartosat (A and F) images at 2.5 m resolution are able to perform well in characterizing high AGB values since the texture-biomass relationship is only subjected to 18 % relative error to that of 15 % in case of IKONOS and could aid in reduction of uncertainty in AGB estimation at a large landscape levels.     

Comparative evaluation of horizontal accuracy of elevations of selected ground control points from ASTER and SRTM DEM with respect to CARTOSAT-1 DEM: a case study of Shahjahanpur district,Uttar Pradesh,India

Digital elevation model (DEM) data of Shuttle Radar Topography Mission (SRTM) are distributed at a horizontal resolution of 90 m (30 m only for US) for the world, Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) DEM data provide 30 m horizontal resolution, while CARTOSAT-1 (IRS-P5) gives 2.6 m horizontal resolution for global coverage. SRTM and ASTER data are available freely but 2.6 m CARTOSAT-1 data are costly. Hence, through this study, we found out a horizontal accuracy for selected ground control points (GCPs) from SRTM and ASTER with respect to CARTOSAT-1 DEM to implement this result (observed from horizontal accuracy) for those areas where the 2.6-m horizontal resolution data are not available. In addition to this, the present study helps in providing a benchmark against which the future DEM products (with horizontal resolution less than CARTOSAT-1) with respect to CARTOSAT-1 DEM can be evaluated. The original SRTM image contained voids that were represented digitally as ?140; such voids were initially filled using the measured values of elevation for obtaining accurate DEM. Horizontal accuracy analysis between SRTM- and ASTER-derived DEMs with respect to CARTOSAT-1 (IRS-P5) DEM allowed a qualitative assessment of the horizontal component of the error, and the appropriable statistical measures were used to estimate their horizontal accuracies. The horizontal accuracy for ASTER and SRTM DEM with respect to CARTOSAT-1 were evaluated using the root mean square error (RMSE) and relative root mean square error (R-RMSE). The results from this study revealed that the average RMSE of 20 selected GCPs was 2.17 for SRTM and 2.817 for ASTER, which are also validated using R-RMSE test which proves that SRTM data have good horizontal accuracy than ASTER with respect to CARTOSAT-1 because the average R-RMSE of 20 GCPs was 3.7 × 10^?4 and 5.3 × 10^?4 for SRTM and ASTER, respectively.     

面向对象的遥感影像模糊分类方法研究

传统的基于像素的遥感影像处理方法都是基于遥感影像光谱信息极其丰富,地物间光谱差异较为明显的基础上进行的。对于只含有较少波段的高分辨率遥感影像,传统的分类方法,就会造成分类精度降低,空间数据的大量冗余,并且其分类结果常常是椒盐图像,不利于进行空间分析。本文采用面向对象的影像分类方法,考虑了对象的不同特征值,例如光谱值,形状和纹理,结合上下文关系和语义的信息,这种分类技术不仅能够使用影像属性,而且能够利用不同影像对象之间的空间关系。在对诸多对象进行分类后,再进行精度分析。在此研究提出了一种面向对象的方法结合模糊理论把许多的对象块分成不同的类别。这一过程主要有两个步骤：第一个步骤是分割。图像分割将整个图像分割成若干个对象,在这个过程中,分割尺度的选择会影响到后续的分类结果和精度。第二个步骤是分类。在这个步骤中,特征值的选择和隶属度函数的选择都对分类结果有着至关重要的影响。