The application of the coalescence clustering algorithm to remotely sensed multispectral data

The coalescence clustering concept of Watanabe has been implemented for the purpose of unsupervised classification of remotely sensed multispectral data. Modifications on the original algorithm were made to enable clustering of limited range discrete data. Application to simulated overlapping Gaussian distributions show that optimal separation of boundaries is achieved at almost every point. Clustering of real data from LANDSAT satellites also yields very meaningful results. Significance of the range parameter and computer requirements are also discussed.     

The estimation of green-leaf-area index from remotely sensed airborne multispectral scanner data

Airborne multispectral scanning system (MSS) data from the Natural Environment Research Council's MSS-82 project were used to estimate the green-leaf-area index (GLAI) of an area of semi-natural and agricultural limestone grassland. This research is the pilot study of a five phase investigation which will ultimately test the feasibility of using airborne and spaceborne MSS data to estimate GLAI over large areas. The pilot study was designed around four stages: (i) derivation of the relationship between multispectral reflectance and vegetation amount using ground radiometric data collected in the field, (ii) production of a perpendicular-vegetation index (PVI) image, (iii) production of a GLAI image using the relationship between PVI and GLAI and (iv) accuracy assessment of the resultant GLAI image. Due to a number of project- and environment-specific problems the proposed methodology was modified and GLAI was predicted to an accuracy of 50-86 per cent at the 95 per cent confidence level. For the future, a change of project design and the use of a vegetation model to correct for environmental anomalies will enable the technique to be used to greater effect.     

Knowledge-based interpretation of remotely sensed images

A review of the current approaches to machine interpretation of remotely sensed data is presented. The types of external knowledge which may be accessed in order to improve interpretation of remotely sensed data are introduced. The techniques which use this external knowledge are then discussed, together with their applicability to the data.     

Cloud removal for hyperspectral remotely sensed images based on hyperspectral information fusion

ABSTRACT

Hyperspectral remote sensing plays an important role in a wide variety of fields. However, its specific application for land surface analysis has been constrained due to the different shapes of thick, opaque cloud cover. The reconstruction of missing information obscured by clouds in remote-sensing images is an area of active research. However, most of the available cloud-removal methods are not suitable for hyperspectral images, because they lose the spectral information which is very important for hyperspectral analysis. In this article, we developed a new spectral resolution enhancement method for cloud removal (SREM-CR) from hyperspectral images, with the help of an auxiliary cloud-free multispectral image acquired at different times. In the fixed hyperspectral image, spectra of the cloud cover pixels are reconstructed depending on the relationship between the original hyperspectral and multispectral images. The final resulting image has the same spectral resolution as the original hyperspectral image but without clouds. This approach was tested on two experiments, in which the results were compared by visual interpretation and statistical indices. Our method demonstrated good performance.     

Distributed computing model for processing remotely sensed images based on grid computing

With advances in remote-sensing technology, the large volumes of data cannot be analyzed efficiently and rapidly, especially with arrival of high-resolution images. The development of image-processing technology is an urgent and complex problem for computer and geo-science experts. It involves, not only knowledge of remote sensing, but also of computing and networking. Remotely sensed images need to be processed rapidly and effectively in a distributed and parallel processing environment. Grid computing is a new form of distributed computing, providing an advanced computing and sharing model to solve large and computationally intensive problems. According to the basic principle of grid computing, we construct a distributed processing system for processing remotely sensed images. This paper focuses on the implementation of such a distributed computing and processing model based on the theory of grid computing. Firstly, problems in the field of remotely sensed image processing are analyzed. Then, the distributed (and parallel) computing model design, based on grid computing, is applied. Finally, implementation methods with middleware technology are discussed in detail. From a test analysis of our system, TARIES.NET, the whole image-processing system is evaluated, and the results show the feasibility of the model design and the efficiency of the remotely sensed image distributed and parallel processing system.     

Unsupervised change detection based on robust chi-squared transform for bitemporal remotely sensed images

Chi-squared transform (CST)-based methods are simple and effective methods for detecting changes in remotely sensed images that have been registered and aligned. The methods operate directly on information stored in the difference image. However, the estimated mean and covariance matrix of the Gaussian distribution that describes the unchanged pixels can be biased when the changed pixels (outliers) are also included. To overcome this issue, we propose a pixel-based unsupervised change detection method that gives robust estimates of these parameters. The method is iterative but requires only a small number of iterations. In addition, we also design an algorithm to automatically search for the optimal threshold that is needed for classifying changed versus unchanged pixels. This algorithm finds the optimal threshold where the mean and covariance matrix of the change detection result most agree with those statistics obtained from the above-mentioned robust algorithm. We refer to our change detection method as the robust CST (RCST) method. The proposed method has been evaluated on two image data-sets and compared with four state-of-the-art methods. The effectiveness of RCST is confirmed by its low overall errors (OE) and high kappa coefficients on both data-sets.     

Restoration of clouded pixels in multispectral remotely sensed imagery with cokriging

The presence of clouds and their shadows in remotely sensed images limits their potential uses for extracting information. The commonly used methods for replacing clouded pixels by land cover reflection estimates usually yield poor results if the images being combined exhibit radical differences in target radiance due, for example, to large date separation and high temporal variability. This study focuses on introducing geostatistical techniques for interpolating the DN values of clouded pixels in multispectral remotely sensed images using traditional ordinary cokriging and standardized ordinary cokriging. Two case studies were conducted in this study. The first case study shows that the methods work well for the small clouds in a heterogeneous landscape even when the images being combined show high temporal variability. Although the basic spatial structure in large size clouds can be captured, image interpolation‐related artefacts such as smoothing effects are visually apparent in a heterogeneous landscape. The second case study indicates that the cokriging methods work better in homogenous regions such as the dominantly agricultural areas in United States Midwest. Various statistics including both global statistics and local statistics are employed to confirm the reliability of the methods.     

Calibrated tree counting on remotely sensed images of planted forests

The development of robust and accurate methods for counting trees from remotely sensed data could provide substantial cost savings in forest inventory. A new methodology that provides a framework for calibrating tree detection algorithms to obtain accurate tree counts for even-aged stands is described. The methodology was evaluated using two tree detection algorithms and two operators using airborne laser scanning (ALS) and orthophotograph images for four Pinus radiata D.Don stands ranging in age between 5 and 32 years with stand densities ranging between 204 and 826 stems ha^?1. For application of the methodology to ALS images the error of estimate on the total count was 4.7% when calibration counts from actual ground plots were used and 10.5% when calibration counts from virtual plots on the image were used. For orthophotographs, the error of estimate was 6.1% using ground calibration plots and 24.3% using calibration counts from virtual plots. The described methodology was shown to be robust to variations in the process from the two operators and two algorithms evaluated. The measure of accuracy determined using the methodology can be used to provide an objective basis for evaluating a wide range of tree counting and detection processes in future research.     

VDM-RS: A visual data mining system for exploring and classifying remotely sensed images

Remotely sensed imagery has become increasingly important in several applications domains, such as environmental monitoring, change detection, fire risk mapping and land use, to name only a few. Several advanced image classification techniques have been developed to analyze such imagery and in particular to improve the accuracy of classifying images in the context of such applications. However, most of the proposed classifiers remain a black box to users, leaving them with little to no means to explore and thus further improve the classification process, in particular for misclassified pixel samples. In this paper, we present the concepts, design and implementation of VDM-RS, a visual data mining system for classifying remotely sensed images and exploring image classification processes. The system provides users with two classes of components. First, visual components are offered that are specific to classifying remotely sensed images and provide traditional interfaces, such as a map view and an error matrix view. Second, the decision tree classifier view provides users with the functionality to trace and explore the classification process of individual pixel samples. This feature allows users to inspect how a sample has been correctly classified using the classifier, but more importantly, it also allows for a detailed exploration of the steps in which a sample has been misclassified. The integration of these features into a coherent, user-friendly system not only helps users in getting more insights into the data, but also to better understand and subsequently improve a classifier for remotely sensed images. We demonstrate the functionality of the system's components and their interaction for classifying imagery using a hyperspectral image dataset.     

Information extraction from remotely sensed data.

Digital image processing is now widely available for users of remotely sensed data. Although such processing offers many new opportunities for the user (or analyst) it also makes heavy demands on the acquisition of new skills, if the data are to yield useful information efficiently. In deciding on the best approach for image classification the user faces a bewildering array of choices, many of which have been poorly evaluated. It is clear, however, that the use of both internal and external contextual information can be of great value in improving classification performance. The ultimate use of information extracted from remote sensing data is strongly affected by its compatability with other geographic data planes. Problems in achieving such compatibility in the framework of automated geographical information systems are discussed. The success of image analysis and classification methods is highly dependent on the relationships between the abilities of sensing systems themselves and the character of the phenomena being studied. This is illustrated by reference to the capabilities of future high resolution satellite systems.     

Moso bamboo forest extraction and aboveground carbon storage estimation based on multi-source remotely sensed images

Using a combination of moso bamboo forest thematic maps derived from Landsat Thematic Mapper (TM) images, field inventory data, and Moderate Resolution Imaging Spectroradiometer (MODIS) normalized difference vegetation index (NDVI) images, moso bamboo forest was extracted using the matched filtering (MF) technique and its aboveground carbon storage (AGC) was then estimated. This study presents a feasible method for extracting large-scale moso bamboo forests and for estimating moso bamboo forest AGC based on low-spatial resolution MODIS images. The results showed that moso bamboo forests in the majority of counties can be accurately estimated between actual area and estimates, with an R ² of 0.8453. The fitted accuracy of the AGC model was high (R ² = 0.491). The prediction accuracy of the AGC model was also evaluated using validation samples collected from Lin'an City, with an R ² and root mean square error prediction of 0.4778 and 3.06 Mg C ha^?1, respectively. The AGC in the majority of counties or cities in Zhejiang Province was between 0 and 15 Mg C ha^?1, and to a certain extent the predicted AGC estimates were close to observed ground truth data and representative of the study area.     

Burn severity estimation from remotely sensed data: Performance of simulation versus empirical models

Burn severity is a key factor in post-fire assessment and its estimation is traditionally restricted to field work and empirical fitting from remotely sensed data. However, the first method is limited in terms of spatial coverage and cost effectiveness and the second is site- and data-specific. Since alternative approaches based on radiative transfer models (RTM) have been usefully applied in retrieving several biophysical plant parameters (leaf area index, water and dry matter content, chlorophyll), this paper has applied the inversion of a simulation model to estimate burn severity in terms of the Composite Burn Index (CBI). The performance of the model inversion method was compared to standard empirical techniques. The study area chosen was a large forest fire in central Spain which occurred in July 2005. The model inversion showed the most accurate estimation for high severity levels (for CBI > 2.7, RMSE = 0.30) and for unburned areas (CBI < 0.5, RMSE = 0). In both methodologies, the error associated to CBI from 0.5 to 2.7 was not acceptable (RMSE > 0.7), because it is higher than 25% of the total range of the index. Finally, burn severity maps from both methods were compared.     

An adaptive spatially constrained fuzzy c-means algorithm for multispectral remotely sensed imagery clustering

This paper presents a novel adaptive spatially constrained fuzzy c-means (ASCFCM) algorithm for multispectral remotely sensed imagery clustering by incorporating accurate local spatial and grey-level information. In this algorithm, a novel weighted factor is introduced considering spatial distance and membership differences between the centred pixel and its neighbours simultaneously. This factor can adaptively estimate the accurate spatial constrains from neighbouring pixels. To further enhance its robustness to noise and outliers, a novel prior probability function is developed by integrating the mutual dependency information in the neighbourhood to obtain accurate spatial contextual information. The proposed algorithm is free of any experimentally adjusted parameters and totally adaptive to the local image content. Not only the neighbourhood but also the centred pixel terms of the objective function are all accurately estimated. Thus, the ASCFCM enhances the conventional fuzzy c-means (FCM) algorithm by producing homogeneous regions and reducing the edge blurring artefact simultaneously. Experimental results using a series of synthetic and real-world images show that the proposed ASCFCM outperforms the competing methodologies, and hence provides an effective unsupervised method for multispectral remotely sensed imagery clustering.     

Parallel unmixing of remotely sensed hyperspectral images on commodity graphics processing units

Hyperspectral imaging instruments are capable of collecting hundreds of images, corresponding to different wavelength channels, for the same area on the surface of the Earth. One of the main problems in the analysis of hyperspectral data cubes is the presence of mixed pixels, which arise when the spatial resolution of the sensor is not enough to separate spectrally distinct materials. Hyperspectral unmixing is one of the most popular techniques to analyze hyperspectral data. It comprises two stages: (i) automatic identification of pure spectral signatures (endmembers) and (ii) estimation of the fractional abundance of each endmember in each pixel. The spectral unmixing process is quite expensive in computational terms, mainly due to the extremely high dimensionality of hyperspectral data cubes. Although this process maps nicely to high performance systems such as clusters of computers, these systems are generally expensive and difficult to adapt to real‐time data processing requirements introduced by several applications, such as wildland fire tracking, biological threat detection, monitoring of oil spills, and other types of chemical contamination. In this paper, we develop an implementation of the full hyperspectral unmixing chain on commodity graphics processing units (GPUs). The proposed methodology has been implemented, using the CUDA (compute device unified architecture), and tested on three different GPU architectures: NVidia Tesla C1060, NVidia GeForce GTX 275, and NVidia GeForce 9800 GX2, achieving near real‐time unmixing performance in some configurations tested when analyzing two different hyperspectral images, collected over the World Trade Center complex in New York City and the Cuprite mining district in Nevada. Copyright © 2011 John Wiley & Sons, Ltd.     

Super-resolution land-cover mapping using multiple sub-pixel shifted remotely sensed images

Super-resolution land-cover mapping is a promising technology for prediction of the spatial distribution of each land-cover class at the sub-pixel scale. This distribution is often determined based on the principle of spatial dependence and from land-cover fraction images derived with soft classification technology. However, the resulting super-resolution land-cover maps often have uncertainty as no information about sub-pixel land-cover patterns within the low-resolution pixels is used in the model. Accuracy can be improved by incorporating supplemental datasets to provide more land-cover information at the sub-pixel scale; but the effectiveness of this is limited by the availability and quality of these additional datasets. In this paper, a novel super-resolution land-cover mapping technology is proposed, which uses multiple sub-pixel shifted remotely sensed images taken by observation satellites. These satellites take images over the same area once every several days, but the images are not identical because of slight orbit translations. Low-resolution pixels in these remotely sensed images therefore contain different land-cover fractions that can provide useful information for super-resolution land-cover mapping. We have constructed a Hopfield Neural Network (HNN) model to solve it. Maximum spatial dependence is the goal of the proposed model, and the fraction maps of all images are constraints added to the energy function of HNN. The model was applied to synthetic artificial images as well as to a real degraded QuickBird image. The output maps derived from different numbers of images at different zoom factors were compared visually and quantitatively to the super-resolution map generated from a single image. The resulting land-cover maps with multiple remotely sensed images were more accurate than was the single image map. The use of multiple remotely sensed images is therefore a promising method for decreasing the uncertainty of super-resolution land-cover mapping. Moreover, remotely sensed images with similar spatial resolution from different satellite platforms can be used together, allowing a fusion of information obtained from remotely sensed imagery.     

Improvement of remotely sensed low spatial resolution images by back-propagated neural networks using data fusion techniques

Abstract

An effect has been observed in the channel 4 output of the NOAA-12 satellite which has not been observed from previous satellites in the NOAA series. Essentially, when sunglint is being observed by channels 1–3, and especially when channel 3 is saturated, the output of channel 4 becomes paralysed, It is believed that this problem has existed throughout the lifetime of this satellite, since the effect was also observed shortly after launch during July 1991.     

Image processing for boundary extraction of remotely sensed data

Image processing algorithms for analysing remotely sensed data are developed. The algorithms proposed in the paper provide means for autoregressive texture modelling and for boundary detection of uniform subimage areas. The boundary detection methods make use of a semicircular entropy operator and of the binary hypothesis testing of the Poisson counting process. The proposed algorithms are applied to the pattern analysis of the isothermal distribution in the oceanic environment.     

Estimating glacier volume loss using remotely sensed images,digital elevation data,and GIS modelling

Glacier retreat has received much attention in the public and scientific community as a sensitive indicator of global warming. Parameters for changes in glacier size include length, area, volume, and mass balance. While measurement of changes in glacier length and area is relatively straightforward using remotely sensed data, estimating changes in volume and mass balance remains a major challenge. The objectives of this research are to (1) develop a new model to describe the nonlinear thinning process of a glacier surface headward from the terminus position and (2) reconstruct historical glacier longitudinal profiles and 3D surfaces based on current glacier longitudinal profile and thinning rate, and estimate glacier volume loss. Using historical terminus positions and Landsat Thematic Mapper (TM), IKONOS, and Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) global digital elevation model (GDEM) data for the Gangotri Glacier in the Himalayas, longitudinal profiles and 3D surfaces of the glacier for 1900, 1935, and 1971 were reconstructed, and the amount and rate of volumetric losses during 1900–1935, 1935–1971, and 1971–2005 were derived. The methods can be used for more detailed study on estimating volume loss of the Himalayan glaciers and other glaciers.     

Study on component temperatures inversion using satellite remotely sensed data

Because of the low spatial resolution of MODIS data, it is important to decompose mixed pixels to retrieve component temperatures using thermal infrared bands. Grasslands with different coverage conditions are prominent in the area under study. Because of the simple vegetation structure, radiation is less influenced by vegetation shade. If the internal structure of the component parts of the mixed pixel is ignored, the total radiation emitted by the mixed pixel is approximately the sum of the radiation emitted by each component part of the pixel, weighted according to the percentage area of each component part. Vegetation/soil component temperatures based on the sub‐pixel scale are inverted using a constrained optimization algorithm—the genetic algorithm. The study not only broadens the application of the linear spectral mixing model but also develops a practical method for component temperatures retrieval from MODIS satellite data. The results provide more precise parameters for estimation of land surface energy balance and evapotranspiration.