Improvement of user’s accuracy through classification of principal component images and stacked temporal images

The classification accuracy of the various categories on the classified remotely sensed images are usually evaluated by two different measures of accuracy, namely, producer’s accuracy (PA) and user’s accuracy (UA). The PA of a category indicates to what extent the reference pixels of the category are correctly classified, whereas the UA of a category represents to what extent the other categories are less misclassified into the category in question. Therefore, the UA of the various categories determines the reliability of their interpretation on the classified image and is more important to the analyst than the PA. The present investigation has been performed in order to determine if there occurs improvement in the UA of the various categories on the classified image of the principal components of the original bands and on the classified image of the stacked image of two different years. We performed the analyses using the IRS LISS III images of two different years, i.e., 1996 and 2009, that represent the different magnitude of urbanization and the stacked image of these two years pertaining to Ranchi area, Jharkhand, India, with a view to assessing the impacts of urbanization on the UA of the different categories. The results of the investigation demonstrated that there occurs significant improvement in the UA of the impervious categories in the classified image of the stacked image, which is attributable to the aggregation of the spectral information from twice the number of bands from two different years. On the other hand, the classified image of the principal components did not show any improvement in the UA as compared to the original images.     

Can regional aerial images from orthophoto surveys produce high quality photogrammetric Canopy Height Model? A single tree approach in Western Europe

Forest monitoring tools are needed to promote effective and data driven forest management and forest policies. Remote sensing techniques can increase the speed and the cost-efficiency of the forest monitoring as well as large scale mapping of forest attribute (wall-to-wall approach). Digital Aerial Photogrammetry (DAP) is a common cost-effective alternative to airborne laser scanning (ALS) which can be based on aerial photos routinely acquired for general base maps. DAP based on such pre-existing dataset can be a cost effective source of large scale 3D data. In the context of forest characterization, when a quality Digital Terrain Model (DTM) is available, DAP can produce photogrammetric Canopy Height Model (pCHM) which describes the tree canopy height. While this potential seems pretty obvious, few studies have investigated the quality of regional pCHM based on aerial stereo images acquired by standard official aerial surveys. Our study proposes to evaluate the quality of pCHM individual tree height estimates based on raw images acquired following such protocol using a reference filed-measured tree height database. To further ensure the replicability of the approach, the pCHM tree height estimates benchmarking only relied on public forest inventory (FI) information and the photogrammetric protocol was based on low-cost and widely used photogrammetric software. Moreover, our study investigates the relationship between the pCHM tree height estimates based on the neighboring forest parameter provided by the FI program.Our results highlight the good agreement of tree height estimates provided by pCHM using DAP with both field measured and ALS tree height data. In terms of tree height modeling, our pCHM approach reached similar results than the same modeling strategy applied to ALS tree height estimates. Our study also identified some of the drivers of the pCHM tree height estimate error and found forest parameters like tree size (diameter at breast height) and tree type (evergreenness/deciduousness) as well as the terrain topography (slope) to be of higher importance than image survey parameters like the variation of the overlap or the sunlight condition in our dataset. In combination with the pCHM tree height estimate, the terrain slope, the Diameter at Breast Height (DBH) and the evergreenness factor were used to fit a multivariate model predicting the field measured tree height. This model presented better performance than the model linking the pCHM estimates to the field tree height estimates in terms of r² (0.90 VS 0.87) and root mean square error (RMSE, 1.78 VS 2.01 m). Such aspects are poorly addressed in literature and further research should focus on how pCHM approaches could integrate them to improve forest characterization using DAP and pCHM. Our promising results can be used to encourage the use of regional aerial orthophoto surveys archive to produce large scale quality tree height data at very low additional costs, notably in the context of updating national forest inventory programs.     

Landsat MSS images of the clean‐up of Lake Chicot

As a result of the 1991 Persian Gulf war, between mid‐January and June 1991, the Persian Gulf was contaminated with an estimated 4 to 6 million barrels of crude oil, released directly into the Gulf from refinement facilities, transhipment terminals, and moored tankers along the coast of Kuwait, and precipitated from oil fire smoke plumes. To assess the environmental impact of the oil, an international team of marine scientists representing 14 nations was assembled under the auspices of the United Nations International Oceanic Commission and the Regional Organization for Protection of the Marine Environment to conduct detailed surveys of the Persian Gulf, the Strait of Hormuz, and the Gulf of Oman, including hydrographic, chemical, and biological measurements. To supplement the field surveys and to serve as an aid in data interpretation, astronauts aboard the Space Shuttle Atlantis photographed water features and coastal habitats in the Persian Gulf during mission STS‐45 (24 March to 02 April 1992). The astronauts collected 111 hand‐held, color photographs of the Gulf (72 70 mm photographs and 39 5‐inch photographs) from an altitude of 296 km (160 n.mi.). The photographs reveal distributions in water turbidity associated with outflow from the Shatt‐al‐Arab and water circulation along the entire coast of Iran and the Strait of Hormuz, coastal wetlands and shallow‐water habitats, and sticks appearing in the sunglint pattern, which appear to be oil.     

The July 11, 1995 Myanmar–China earthquake: A representative event in the bookshelf faulting system of southeastern Asia observed from JERS-1 SAR images

On July 11, 1995, an Mw 6.8 earthquake struck eastern Myanmar near the Chinese border; hereafter referred to as the 1995 Myanmar–China earthquake. Coseismic surface displacements associated with this event are identified from JERS-1 (Japanese Earth Resources Satellite-1) SAR (Synthetic Aperture Radar) images. The largest relative displacement reached 60 cm in the line-of-sight direction. We speculate that a previously unrecognized dextral strike-slip subvertical fault striking NW–SE was responsible for this event. The coseismic slip distribution on the fault planes is inverted based on the InSAR-derived deformation. The results indicate that the fault slip was confined to two lobes. The maximum slip reached approximately 2.5 m at a depth of 5 km in the northwestern part of the focal region. The inverted geodetic moment was approximately Mw = 6.69, which is consistent with seismological results. The 1995 Myanmar–China earthquake is one of the largest recorded earthquakes that has occurred around the “bookshelf faulting” system between the Sagaing fault in Myanmar and the Red River fault in southwestern China.     

An application of spatial filtering techniques for land cover mapping using TM‐images

Land cover classification using satellite imagery is commonly based on spectral information in the individual pixels. The information in neighbouring pixels is ignored. Spatial filtering techniques using information present in neighbouring pixels may however, contribute significantly to an improvement of the classification. In this study different methods of spatial filtering are applied to a part of a TM‐scene of Kenya to assess their relative reliability. The study area is characterized by extended, relatively homogeneous areas of eucalyptus forests and tea estates and by fragmentated areas of agricultural land use. Spectral information was combined with the results of different spatial filtering methods and then classified. The spatial filtering techniques applied were texture calculation by means of variance, “median minus original” filtering and fractal dimension computations using several sizes of templates. The obtained classification accuracy of several image combinations is compared using the percentage correctly classified and using an overall accuracy measure: the Kappa coefficient. It is concluded that in this case the spatial filtering techniques only slightly improve the classification. From the applied filtering methods texture calculation by means of variance yielded the best results.     

Research on auto-classification method of remote sensing images in mountainous areas—An application in Southwest of China

In mountainous areas, it is the undulant terrain, various types of geomorphic and land use that make the remote sensing images great metamorphism. Moreover, due to the elevation, there are many areas covered with shadow, clouds and snow that make the images more inaccurate. As a result, it would be very difficult to carry out auto-classification of RS images in these areas. The study took Southwest China as the case study area and the TM images, SPOT images as the basic information sources assisted by the auxiliary data of DEM, NDVI, topographical maps and soil maps to preprocess the images. After preprocessing by topographic correction and wiping off clouds, snow and shadows, all the image data were stacked together to form the images to be classified. Then, the research used segmentation technology and hierarchical method to extract the main types of land use in the area automatically. The results indicated that the qualitative accuracies of all types of land use extracted in Southwest China were above 90 percent, and the quantitative accuracies was above 86 percent. The goal of reducing workloads had been realized. Supported by the National Public Welfare Project on Environmental Protection (2007KYYW21), the Program of National Science and Technology research( 2006BAC01A01-05).     

Spectral indicators of forage quality in West Africa’s tropical savannas

Forage is among the essential ecosystem services provided by tropical savannas. Expected changes in climate and land use may cause a strong decline in herbaceous forage provision and thus make it advisable to monitor its dynamics. Spectroscopy offers promising tools for fast and non-destructive estimations of forage variables, yet suffers from unfavourable measurement conditions during the tropical growing period such as frequent cloud cover and high humidity. This study aims to test whether spatio-temporal information on the quality (metabolisable energy content, ME) and quantity (green biomass, BM) of West African forage resources can be correlated to in situ measured reflectance data. We could establish robust and independent models via partial least squares regression, when spectra were preprocessed using second derivative transformation (ME: max. adjusted R² in validation (_adjR²_VAL) = 0.83, min. normalised root mean square error (nRMSE) = 7.3%; BM: max. _adjR²_VAL = 0.75, min. nRMSE = 9.4%). Reflectance data with a reduced spectral range (350–1075 nm) still rendered satisfactory accuracy.Our results confirm that a strong correlation between forage characteristics and reflectance of tropical savanna vegetation can be found. For the first time in field spectroscopy studies, forage quality is modelled as ME content based on 24-h in vitro gas production in the Hohenheim gas test system and crude protein concentration of BM. Established spectral models could help to monitor forage provision in space and time, which is of great importance for an adaptive livestock management.     

Early-season mapping of crops and cultural operations using very high spatial resolution Pléiades images

The aim of this study was to assess the contribution of very high spatial resolution (VHSR) Pléiades images to both early season crop identification and the mapping of bare soil surface characteristics due to cultural operations. The study region covering 21 km² is located west of the peri-urban territory of the Versailles plain and the Alluets plateau (Yvelines, France). About 100 cropped fields were observed on the ground synchronously with two Pléiades images of 3 and 24 April 2013 and one SPOT4 image of 2 April 2013. The GIS structuring of these field data along with vector information about field boundaries was used for delimitating both training and test zones for the support vector machine classifier with polynomial function kernel (pSVM). The pSVM was computed on the spectral bands and NDVI for both single-date Pléiades and the bi-temporal Pléiades pair. For the single-date classifications of crops, the overall per-pixel accuracy reached 87% for the SPOT4 image of 2 April (6 classes), 79% for the Pléiades image of 3 April (6 classes) and 82% for that of 24 April (7 classes). At the earlier date (2–3 April), the Pléiades image very well discriminated cultural operations (>77%, user’s or producer’s accuracies) as well as fallows and grasslands, while winter cereals and rapeseed were better discriminated by the SPOT4 image winter cereals (>70%, user’s or producer’s accuracies). As Pléiades images revealed within-field spatial variations of early phenological stages of winter cereals that could be critical for adjusting management of zones with delayed development during the growing season, they brought information complementary to multispectral images with high spatial resolution. For the bi-temporal Pléiades image, the overall per-pixel accuracy was about 80% (7 classes), winter crops, grasslands and fallows being very well detected while confusions occurred between spring barley at initial stages (2–3 leaves) and bare soils prepared for other spring crops. Using an additional validation field set covering ∼1/3 of the study area croplands, the crop map resulting from the bi-temporal Pléiades pair achieved correct crop prediction for about 89.7% of the validation fields when considering composite classes for winter cereals and for spring crops. Early-season Pléiades images therefore show a considerable potential for anticipating regional crop patterns and detecting soil tillage operations in spring.     

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%.     

The potential of Pléiades images with high angle of incidence for reconstructing the coastal cliff face in Normandy (France)

To monitor chalk cliff face along the Normandy coast (NW France) which is prone to erosion, we tested the potential of cliff face 3D reconstruction using pairs of images with high angle of incidence at different dates from the agile Pléiades satellites. The verticality aspect of the cliff face brings difficulties in the 3D reconstruction process. Furthermore, the studied area is challenging mainly because the cliff face is north-oriented (shadow). Pléiades images were acquired over several days (multi-date stereoscopic method) with requested incidence angles until 40°. 3D reconstructions of the cliff face were compared using two software: ASP® and ERDAS IMAGINE®. Our results are twofold. Firstly, despite ASP® provides denser point clouds than ERDAS IMAGINE® (an average of 1.60 points/m² from 40° incidence angle stereoscopic pairs on the whole cliff face of Varengeville-sur-Mer against 0.77 points/m² respectively), ERDAS IMAGINE® provides more reliable point clouds than ASP® (precision assessment on the Varengeville-sur-Mer cliff face of 0.31 m ± 2.53 and 0.39 m ± 4.24 respectively), with a better spatial distribution over the cliff face and a better representation of the cliff face shape. Secondly, the quality of 3D reconstructions depends mostly on the amount of noise from raw images and on the shadow intensity on the cliff face (radiometric quality of images).     

A spatial–temporal Hopfield neural network approach for super-resolution land cover mapping with multi-temporal different resolution remotely sensed images

The mixed pixel problem affects the extraction of land cover information from remotely sensed images. Super-resolution mapping (SRM) can produce land cover maps with a finer spatial resolution than the remotely sensed images, and reduce the mixed pixel problem to some extent. Traditional SRMs solely adopt a single coarse-resolution image as input. Uncertainty always exists in resultant fine-resolution land cover maps, due to the lack of information about detailed land cover spatial patterns. The development of remote sensing technology has enabled the storage of a great amount of fine spatial resolution remotely sensed images. These data can provide fine-resolution land cover spatial information and are promising in reducing the SRM uncertainty. This paper presents a spatial–temporal Hopfield neural network (STHNN) based SRM, by employing both a current coarse-resolution image and a previous fine-resolution land cover map as input. STHNN considers the spatial information, as well as the temporal information of sub-pixel pairs by distinguishing the unchanged, decreased and increased land cover fractions in each coarse-resolution pixel, and uses different rules in labeling these sub-pixels. The proposed STHNN method was tested using synthetic images with different class fraction errors and real Landsat images, by comparing with pixel-based classification method and several popular SRM methods including pixel-swapping algorithm, Hopfield neural network based method and sub-pixel land cover change mapping method. Results show that STHNN outperforms pixel-based classification method, pixel-swapping algorithm and Hopfield neural network based model in most cases. The weight parameters of different STHNN spatial constraints, temporal constraints and fraction constraint have important functions in the STHNN performance. The heterogeneity degree of the previous map and the fraction images errors affect the STHNN accuracy, and can be served as guidances of selecting the optimal STHNN weight parameters.     

Evaluation and Comparison of Multi Resolution DEM Derived Through Cartosat-1 Stereo Pair – A Case Study of Damanganga Basin

The study evaluates and compares Digital Elevation Model (DEM) data of various grid spacing derived using high resolution Cartosat 1 stereo data for hydrologic applications. DEM is essential in modeling different environmental processes which depend on surface elevation. The accuracy of derived DEM varies with grid spacing and source. The CartoDEM is the photogrammetric DEM derived from stereo pairs. Damanganga basin lying in the Western Ghats was analysed using 11 Carto stereo pairs. The process of triangulation resulted in RMSE of 0.42. DEM was extracted at 10 m, 20 m, 30 m, 40 m, 50 m and 90 m grid spacing and compared with ASTER GDEM (30 m) and SRTM DEM (90 m). DEM accuracy was checked with Root Mean Square Error (RMSE) statistic for random points generated in different elevation zones. Extracted stream networks were compared based on Correctness Index and Figure of Merit index, calculated for all the Digital Elevation Models at varying cell sizes. In order to further evaluate the DEM’s, a simple flood simulation with no water movement and no consideration of real time precipitation data was carried out and relationship between heights of flood stage and inundation area for each Digital Elevation Model was also established.     

A Review of the Policies on the Private Colleges of China

One of the most impor tant factors impacting the development of today‘s private higher education in China is that there are not enough policies to support in.Then,the history of private higher educationin Nanjing Nationalist Government(1927-1749)is focused on.     

基于Sentinel-1A数据的土壤水分反演

土壤水分是影响作物生长的重要因素,也是监测旱情、估算作物产量的重要参量.为及时、准确地掌握土壤水分,在利用水云模型(Water Cloud Model,WCM)对Sentinel-1 A的后向散射系数校正的基础上,联合地面土壤水分数据,采用线性回归、BP神经网络和支持向量回归三类模型进行了地表土壤水分反演实验研究.实验...     

A Test of Central Place Theory Using Shuttle Imaging Radar (SIR‐A) of China's North Central Plain

Abstract

This study utilizes radar imagery to test the application of Walter Christaller's central‐place theory on China's North Central Plain. Examination of the spatial order and layout of settlements reveal that central place theory's transportation principle, k=4 exists. The study distinguishes four hierarchical settlement sizes and shows how effective rural societies are in evolving settlement structures whose spatial order and regularity avoids spatial inefficiencies. The advent of space based imaging systems such as SIR‐A radar imagery has permitted tests of new sites where the restrictive conditions of central place theory exist. These studies allow parallels to be drawn between the postulates of this theory of settlement structure and whether the theory is matched in reality.

This work is needed because of such scarcities. It allows the theory to undergo further subjections, a scrutiny needed for the validation, or otherwise of its postulates. The work adds to a large body of literature on this topic whose limitation has been the scarcity of actual test sites where theory and reality can be compared. The study presents positive indications that central place theory exists in reality, a compliment to the remarkable perceptions of Walter Christaller.     

利用HJ-1A/1BCCD数据进行黄河凌汛监测

利用HJ-1A/1B卫星CCD数据进行黄河凌汛监测,提出了基于主成分分析和决策树的冰凌提取方法。实验结果显示,利用HJ-1A/1B卫星CCD数据能有效地提取出黄河的冰凌范围,对黄河凌汛监测具有较大的应用潜力。     

How many samples are needed? An investigation of binary logistic regression for selective omission in a road network

ABSTRACT

Selective omission in a road network (or road selection) means to retain more important roads, and it is a necessary operator to transform a road network at a large scale to that at a smaller scale. This study discusses the use of the supervised learning approach to road selection, and investigates how many samples are needed for a good performance of road selection. More precisely, the binary logistic regression is employed and three road network data with different sizes and different target scales are involved for testing. The different percentages and numbers of strokes are randomly chosen for training a logistic regression model, which is further applied into the untrained strokes for validation. The performances of using the different sample sizes are mainly evaluated by an error rate estimate. Significance tests are also employed to investigate whether the use of different sample sizes shows statistically significant differences. The experimental results show that in most cases, the error rate estimate is around 0.1–0.2; more importantly, only a small number (e.g., 50–100) of training samples is needed, which indicates the usability of binary logistic regression for road selection.