Search

An important system-performance parameter for remote-sensing systems is the spatial resolution. This term has an imprecise meaning and there are a multitude of different definitions. In this paper, the most important definitions are reviewed and consideration is given to the factors which influence the resolution of a given system. It is concluded that no single definition of spatial resolution can be applied to all systems, nor can spatial resolution alone be used to measure the significant information content of image data.

With the rapid development of remote sensing technologies, the frequency of observations of the same location is increasing, and many satellites and sensors produced a large amount of time series images. These images make long-term change detection and dynamic characteristic estimation of ground features possible. However, conventional remote sensing image change detection methods mostly rely on manual visual interpretation and supervised or unsupervised computer-aided classification. Traditional methods always face many bottlenecks when processing big and fast-growing datasets, such as low computational efficiency, low level of automation, and different identification standards and accuracies caused by different operators. With the rapid accumulation of remote sensing data, it has become an important but more challenging task to conduct change detection in a more precise, automated and standardized way. The development of geointelligent computing technologies provides a means of solving these problems and improve the accuracy and efficiency of remote sensing image change detection. In this paper, we presented a novel deep learning model called nest network(NestNet) based on a convolutional neural network to improve the accuracy of the automatic change detection task by using remotely sensed time series images. NestNet extracts the respective features of bi-temporal images using an encoding parallel module and subsequently employs absolute different operations to process the features of two images. Compared with change detection method based on U-Shaped network plus plus (UNet++), the parallel module improves the efficiency of NestNet. Finally, a decoding module is used to generate a predicted change image. This paper compares NestNet to traditional methods and state-of-the-art deep learning models on two datasets. The experimental results demonstrate that the accuracy of NestNet is better than that of state-of-the-art methods. It can be concluded that the NestNet model is a potential approach for change detection using high resolution remote sensing images.

International audience; It is a challenge to develop methods which can process the PolSAR and multispectral (MS) data modalities together without losing information from either for remote sensing applications. This paper presents a study which attempts to introduce novel deep learning based remote sensing data processing frameworks that utilizes convolutional neural networks (CNNs) in both spatial and spectral domains to perform land cover (LC) classification with PolSAR-MS data. Also since earth observation remotely sensed data have usually larger spectral depth than normal camera image data, exploiting the spectral information in remote sensing (RS) data is crucial as well. In fact, convolutions in the sub-spectral space are intuitive and alternative to the process of feature selection. Recently, researchers have gained success in exploiting the spectral information of RS data, especially the hyperspectral data with CNNs. In this paper, exploitation of the spectral information in the PolSAR-MS data via a permuted localized spectral convolution along with localized spatial convolution is proposed. Further, the study in this paper also establishes the significance of performing permuted localized spectral convolutions over non-localized or localized spectral convolutions. Two models are proposed, namely a permuted local spectral convolutional network (Perm-LS-CNN) and a permuted local spectral-spatial convolutional network (Perm-LSS-CNN). These models are trained on ground truth class data points measured directly on the terrain. The evaluation of the generalization performance is done using ground truth knowledge on selected well known regions in the study areas. Comparison with other popular machine learning classifiers shows that the Perm-LSS-CNN model provides better classification results in terms of both accuracy and generalization.

The location of building boundary is a crucial prerequisite for geographical condition monitoring, urban management, and building reconstruction. This paper presents a framework that employs a series of algorithms to automatically extract building footprints from airborne light detection and ranging lidar data and image. Connected operators are utilized to extract building regions from lidar data, which would not produce new contours nor change their position and have very good contour-preservation properties. First, the building candidate regions are separated from lidar-derived digital surface model DSM based on a new method proposed within this paper using connected operators, and trees are removed based on the normalized difference vegetation index NDVI value of image. Then, building boundaries are identified and building boundary lines are traced by ‘sleeve’ line simplification method. Finally, the principal directions of buildings are used to regularize the directions of building boundary lines. International Society for Photogrammetry and Remote Sensing ISPRS data sets in Vaihingen whose point spacing is about 0.4 m from urbanized areas were employed to test the proposed framework, and three test areas were selected. A quantitative analysis showed that the method proposed within this paper was effective and the average offset values of simple and complex building boundaries were 0.2–0.4 m and 0.3–0.6 m, respectively.

Synthetic aperture radar (SAR) is a well-proven remote-sensing technique; however, current single-antenna SAR systems cannot fulfil the increasing demands for high-resolution and wide-swath imaging. This paper proposes a multiaperture antenna with waveform diversity for wide-swath remote sensing. This approach employs a multiple-transmit multiple-receive antenna configuration in elevation and an orthogonal transmit waveform. In this way, multiple pairs of virtual beams directed to different subswaths can then be formed simultaneously. Equivalently, a large swath can be obtained. Furthermore, orthogonal frequency diversion multiplexing (OFDM) linearly frequency modulated (LFM) waveform and beamforming on reception are employed in this paper. The system scheme, signal model, processing algorithm, signal-to-noise ratio (SNR), and ambiguity-to-signal ratio (ASR) performance, including both azimuth and range dimensions, are investigated. Simulation results show that the proposed method can obtain an improved r...

This paper presents a few of the research tools of modern physics that can successfully be used to achieve a better understanding of the temporal and spatial variability of the multi-component climate system. The observations obtained by the available remotely sensed instrumentation have led to new advances in climate science, which in conjunction with those reported in the IPCC Fourth Assessment Report constitute the background of the existing knowledge on this subject and are discussed in the present special issue. There is some discussion of matters arising since the IPCC Fourth Assessment Report was published. This paper provides the general background to the papers in this special issue of the journal on remote sensing and climate change.

This paper focuses on the removal of azimuth ambiguities of sparse apertures in distributed microsatellite radar systems. A uniform linear array is ideal for removing azimuth ambiguities. However, the effects of the elliptical orbit and the Earth's rotation make it difficult for conventional space-time processing to resolve Doppler ambiguities. In this paper, we assume that the instantaneous position vector of each satellite is known, and propose a subaperture coordinate-based method. The full aperture geometry is divided into several subaperture independent geometries, in each of which a Cartesian coordinate is constructed according to the position vector of a reference satellite. After phase error removal the baselines in each subaperture coordinate system are obtained by numerical calculation. For each subaperture's returned data, we first compensate for the phase error caused by the three-dimensional (3D) sparse arrays, and then transform the long-track sparse arrays to short along-track arrays. Using a beamforming technique, the Doppler ambiguity is suppressed for each subaperture followed by combining them into the full aperture and full bandwidth spectrum. Theoretical derivation, performance analysis and simulation results are presented.

Most of the research in remote sensing is focused on developing a well-defined and reliable automated process for the extraction of information from different types of imagery. This paper deals with the extraction of linear anthropogenic objects (especially buildings) in hilly areas. The analysis of digital surface models from hilly terrain data is still challenging, especially for highly sloped landscapes. Commonly, the first task to study satellite data is to separate ground and object points as a preparatory step for further object classification. In this paper, object and ground separation in hilly terrain (PAN image of Indian Remote Sensing (IRS)-1C satellite) is addressed by employing wavelets. Then the extraction of desired objects (buildings) is done by employing the marker-controlled watershed segmentation. Experimental results demonstrate the method to be effective.

This paper focuses on the establishment of a pixel-level fusion framework for optical and synthetic aperture radar (SAR) images to combine these two types of remotely sensed imagery for feature enhancement. We have proposed a new fusion technique, namely block-based synthetic variable ratio (Block-SVR), which is a technique based on multiple linear regression of block regions to fuse optical and SAR imagery. In order to investigate the effectiveness of the method, the fusion results of a higher resolution airborne SAR image and a lower resolution multispectral image are presented. According to the fusion results, the fused images have enhanced certain features, namely the spatial and textural content and features that are invisible in multispectral images, while preserving colour characteristics. The spectral, spatial and textural effects of the presented algorithm were also evaluated mainly by visual and quantitative methods, and compared to those of intensity-hue-saturation (IHS), principal component analysis (PCA) and wavelet-based methods. During the implementation of the block-regression based technique there are at least two advantages. One is that the block-regression based technique drastically decreases the amount of computation, whereas regression of the whole scene image is almost impossible. The other, most important, advantage is that adjustment of regressed block size can result in different emphasis between preservation of spectral characteristics and enhancement of spatial and textural content. The larger the regression block, the more the spatial and textural details are enhanced. In contrast, the smaller the regression block, the more the spectral features are preserved. The assessments indicate that the block-regression based method is more flexible than others, because it can achieve a satisfactory trade-off between preservation of spectral characteristics and enhancement of spatial and textural content by selection of optimal block size with respect to visual interpretation and mapping. This paper also proposes a scheme for the fusion of SPOT5 panchromatic, XS images with airborne SAR images using the block-regression based technique.

Many factors influence the quality and value of a classification accuracy assessment and evaluation programme. This paper focuses on the size of the testing set(s) used with particular regard to the impacts on accuracy assessment and comparison. Testing set size is important as the use of an inappropriately large or small sample could lead to limited and sometimes erroneous assessments of accuracy and of differences in accuracy. Here, some of the basic statistical principles of sample size determination are outlined, including a discussion of Type II errors and their control. The paper provides a discussion on some of the basic issues of sample size determination for accuracy assessment and includes factors linked to accuracy comparison. With the latter, the researcher should specify the effect size (minimum meaningful difference in accuracy), significance level and power used in an analysis and ideally also fit confidence limits to derived estimates. This will help design a study and aid the use of appro...

Polar orbiting meteorological satellites have been used successfully in many fields in which applications, such as dynamic monitoring, need multi-temporal remote sensing data imaged in the same region but at different times and with different sensors. The basis of these quantitative applications is automatic and synthetic preprocessing of different satellite data. In order to ensure the direct comparability of multi-source satellite National Oceanic & Atmospheric Administration-Advanced Very High Resolution Radiometer (NOAA-AVHRR) series and FY-1D 1A.5 data, based on its imaging mechanisms, the paper discusses radiant and geometric preprocessing methods used for visual and near-infrared and thermal infrared channels. This is performed by analysing the technical parameters of the satellites and the sensors, theoretical numerical simulation and image data comparison. For synthetic radiance preprocessing, radiance calibration, spectral calibration for different sensors and calibration of the Sun elevation angles or the satellite zenith angles at the imaging time are included. Based on normalization of spatial resolution in scan-line directions, rough geometric correction and precise registration are sequentially processed for geometric preprocessing with a higher precision. Combining the detailed steps introduced in the paper, automatic preprocessing of polar orbiting meteorological satellites can be realized.

In clustering analysis of remote sensing imagery, a commonly held assumption is that the feature space can be modelled as a mixture of Gaussians. However, the assumption is not true for many real data and therefore incorrect classification results are often obtained by parametric methods. Nonparametric methods in feature space analysis can avoid the use of the normality assumption. Arbitrarily structured feature spaces can be analysed only by means of nonparametric methods as these methods do not have embedded assumptions. The mean shift is a basic computational module of the nonparametric technique in pattern recognition. The mean shift procedure can be used to cluster multispectral remote sensing imagery. Earlier clustering techniques based on the mean shift used a single scale over the entire feature space and were not feasible for the analysis of complex multimodal feature spaces. In this paper, we present an adaptive mean shift method in which local scale information is involved. The proposed algorithm can find arbitrary density, size and shape clusters in remote sensing imagery. The method is a simple technique of unsupervised image classification. We demonstrate its advantages in classification accuracy over earlier methods described in this paper.

The detection of surfaces cultivated with coca plants is essential to all activities related to monitoring and combating the production of illicit drugs such as cocaine. Moreover, deforestation processes can often be associated with the presence of coca plantations in the Andean region. In this context, the use of remote sensing technologies has proven to be effective for building an operational monitoring capacity, but more effort is needed to establish a robust computational frame for automatic detection procedures. This paper contributes to this effort by analysing the discrimination power of different textural measures and by proposing a new Mature Coca Index (MCI) derived from the textural analysis of 1-m resolution satellite imagery. The approach can be applied to the latest generation optical sensors such as IKONOS and QUICKBIRD, and improves the current state-of-the-art of recognition procedures for satellite data based on radiometric discrimination. The paper describes the MCI concept, its implementation, and application to a data sample derived from the IKONOS sensor.

In this paper we present the results of time series analysis for a coal mining region based on partial unmixing. We test the method also known as mixture tuned matched filtering on an eight image Landsat 5 TM and Landsat 7 ETM+ time series covering the period from 1987 to 2003. Common change detection methods often include the comparison of two interactively generated classification results, such as derived from Maximum Likelihood classification. These approaches often yield highly accurate results. However, disadvantages include a strong analyst bias and hardly repeatable results. For a quantitative monitoring of a single surface class' development over time they are often not recommendable. Our goal was to test an unbiased quantitative way to assess the development of coal surfaces, such as outcropping coal seams, coal storage piles, coal waste piles, and coal washery discard, within multiple date satellite imagery. Partial unmixing approaches were developed to detect one or few target materials surrounded by—or mixed with—an unknown background material. The main advantage is that only the spectral characteristics of the material of interest must be known, and the desired material can furthermore occur with subpixel coverage. Crisp pixel classificators like maximum likelihood on the other hand require knowledge of all classes. They can only map materials which dominate a pixel. Linear unmixing procedures such as partial unmixing require a thorough radiometric pre-processing of data. Furthermore, the accuracy and representativity of selected input spectra must be granted. In this paper we demonstrate that partial unmixing is a powerful method to detect and extract single landcover classes of interest relatively fast and unbiased. The subpixel fraction percentages should be interpreted in a relative way only. We furthermore show that partial unmixing represents a standardized method for time series analyses and allows for a quantitative assessment of the temporal development of an area. Challenges lie in the validation of partial unmixing results, which we realized through thresholding of unmixing results and accuracy assessment with ground truth polygons mapped in situ. Furthermore, we performed an indirect comparison with results of a multi-endmember unmixing.

Compared to remote sensing images of medium or low spatial resolution, high-resolution remote sensing images can provide observation data containing more detailed information for georesearch. Accordingly, an important issue for current computer and geoscience experts is to develop useful methods or technology to extract information from these high-resolution satellite images. As part of a series of research into object extraction, this paper focuses mainly on the extraction of bridges over water from high-resolution panchromatic satellite images. Since bridges over water are obviously adjacent to water in remote sensing images, this paper proposes a practical knowledge-based bridge extraction method for remote sensing images of high spatial resolution. The steps involved are: water extraction based on Gauss Markov Random Field (GMRF)-Support Vector Machine (SVM) classification methods which use a SVM to classify the image based on textural features expressed by a GMRF; image thinning and removal of fragmented lines; main trunk detection by width; vectorization; and feature expression. Finally, tests are described for two pieces of panchromatic IKONOS satellite images with a 1 m resolution. The experimental results show that the proposed method is suitable for images with a single-peak histogram (contrast between water and land is sharp) or a multi-peak histogram (greyscale value of water is close to that of land).

Based on high-resolution SAR data, in this paper, a novel automatic matching model is proposed. The model, which employs a coarse to fine strategy as a whole, consists of three steps. In the first step, edge features are extracted on different levels of pyramid images and an efficient Hausdorff distance-based method is used to yield a coarse global feature match. Due to bi-tree searching, the bottleneck of Hausdorff distance's matching is well resolved. Secondly, SSDA (Sequence Similarity Detection Algorithm) is employed to acquire tie-points using a cross-searching approach which treats features extracted from master and slave images equally. Finally, local-adaptive splitting algorithm with MMSE (Minimum Mean Square Error) is used to achieve a fine matching; local-adaptive splitting algorithm is the essential process to achieve sub-pixel matching accuracy, which enhances the process's flexibility and robustness. Airborne SAR images with high resolution are provided by the Institute of Electronics, CAS and used for experiments-the results of the experiments demonstrate that the model proposed in this paper is robust, with high accuracy (up to a fraction of a pixel), and can be successfully applied to automatic matching of high-resolution SAR images.

All Sun-synchronous remote sensing satellites, during day passes always acquire images when the illumination source (i.e. Sun position) is from the SE direction. The typical solar-illumination-source-observer position creates a false topographic perception phenomenon (FTPP) in the images of any rugged terrain (e.g. Himalayas, Alps etc.). Due to the presence of FTPP, valleys appear as ridges and vice versa, especially in rugged terrain. The correction of FTPP is necessary because it creates confusion in the minds of interpreters due to inherent inverse topographic perception. This paper explains the development of a new approach to remove FTPP. It has been successfully demonstrated using an IRS-1D-LISS-III image of part of the Himalayas, which represents highly rugged terrain. In this newly developed image-processing technique; a red-green-blue (RGB) image is first transformed into intensity-hue-saturation (IHS) channels, then the original intensity channel is inverted and later retransformed back from an IHS to a RGB image. After this retransformation, the RGB image is then colour balanced with original false colour composite. The final image product is free from FTPP and also contains almost the same image information. This method is an easy and quicker method to remove FTPP in comparison to the already available methods (e.g. image rotation by 180°; creating an image negative; using an opposite illuminated shaded relief model as an intensity image during IHS to RGB retransformation) for FTPP correction. Further, this paper discusses the relative advantages of this new method in comparison to previous methods in light of image classification and lineament interpretation carried out using images with and without FTPP.

This paper is an attempt to suggest an approach for eliminating the lengthy process of selecting various factors while using back‐propagation artificial neural network (BPANN) and to quantify the relative importance of the factors affecting classification results. A novel approach called conjoint analysis has been used here. The paper also presents the classification results of an Indian urban environment using two BPANN approaches and compares them with the conventional Gaussian maximum likelihood (GML) classification approach. The study showed that conjoint analysis can be successfully used to select various parameters of BPANN prior to carrying out the classifications using any of the BPANN approach. Factors like size of training samples and first hidden layer come out as some of the most important factors while the second hidden layer has the least affect on classification accuracy. The resilient back‐propagation method of BPANN is the best and robust method for urban classification. Results also show...

The increased synergy between neural networks (NN) and fuzzy sets has led to the introduction of granular neural networks (GNNs) that operate on granules of information, rather than information itself. The fact that processing is done on a conceptual rather than on a numerical level, combined with the representation of granules using linguistic terms, results in increased interpretability. This is the actual benefit, and not increased accuracy, gained by GNNs. The constraints used to implement the GNN are such that accuracy degradation should not be surprising. Having said that, it is well known that simple structured NNs tend to be less prone to over-fitting the training data set, maintaining the ability to generalize and more accurately classify previously unseen data. Standard NNs are frequently found to be accurate but difficult to explain, hence they are often associated with the black box syndrome. Because in GNNs the operation is carried out at a conceptual level, the components have unambiguous meaning, revealing how classification decisions are formed. In this paper, the interpretability of GNNs is exploited using a satellite image classification problem. We examine how land use classification using both spectral and non-spectral information is expressed in GNN terms. One further contribution of this paper is the use of specific symbolization of the network components to easily establish causality relationships.

Edge detection of an image by remote sensing is an important tool for the detection of oil spilled on sea. There are many algorithms for edge detection by image processing. Each algorithm has its own advantages and disadvantages for different images. In this paper, the author makes some improvements to the Pal–King fuzzy edge detection algorithm and proposes an algorithm combining improved fuzzy theory and a genetic algorithm for the detection of oil spilled on the sea by remote sensing. The Pal–King fuzzy detection algorithm has a good capability for the detection of fuzzy and thin edges. However the complex and large calculation and the fixed threshold value do not suit some kinds of image processing. The algorithm presented in this paper is composed of two parts: an improved fuzzy enhancement algorithm, which simplifies the complex G and G −1 calculation in the Pal–King algorithm; a genetic algorithm with which we are able to obtain the threshold value precisely and quickly for image processing. Finall...

Natural and anthropogenic processes at the Earth's surface operate at a range of spatial and temporal scales. Different scales of observation are required to match the spatial scales of the processes under observation. At the same time, the temporal sampling rate of the observing systems must be reconciled with the dynamics of the processes observed. Bringing together these issues requires insight, innovation and, inevitably, compromise. This paper reviews spatial and temporal considerations in remote sensing and introduces the papers in this Special Issue on ‘Scales and Dynamics in Observing the Environment’. The review comprises three main sections. The first section focuses on spatial variability in remote sensing, while the second section focuses on temporal variability in remote sensing. The third section links these two issues, focusing on the interplay of space and time in remote sensing. The review is primarily theoretical, explaining spatial and temporal properties of remote sensing and remotely ...

Land cover classification over large geographic areas using remotely sensed data is increasingly common as a result of the requirements of national inventory and monitoring programmes, scientific modelling and international environmental treaties. Although large-area land cover products are more prevalent, standard operational protocols for their validation do not exist. This paper provides a framework for the accuracy assessment of large-area land cover products and synthesizes some of the key decision points in the design and implementation of an accuracy assessment from the literature. The fundamental components of a validation plan are addressed and the framework is then applied to the land cover map of the forested area of Canada that is currently being produced by the Earth Observation for Sustainable Development programme. This example demonstrates the compromise between the theoretical aspects of accuracy assessment and the practical realities of implementation, over a specific jurisdiction. The framework presented in this paper provides an example for others embarking on the assessment of large-area land cover products and can serve as the foundation for planning a statistically robust validation.

Moving objects have changing and repeating patterns due to their movements in space over time. A set of observations on a moving object can be used to derive the movement patterns by summarizing frequently visited locations and frequently used paths over time by the object. These space- and time-dependent movement patterns of an object are referred to as spatio-temporal movement signatures in this paper. There is a need to understand and identify the spatio-temporal movement signatures. Analysis of the spatio-temporal movement signatures can help improve the modelling, querying, and reasoning capabilities of the movements of objects. This paper demonstrates the extraction of movement signatures from sets of movement observations using Visualization and Resilient Neural Network methodologies. Identification of movement signatures and definition of their attributes provide summary-level information for modelling and reasoning about moving objects.

This paper presents an improvement of epipolar plane image (EPI) analysis technique for producing high‐accuracy urban 3D surface data from remotely sensed airborne image sequences. The method first forms 3D spatio‐temporal solid data from the image sequences, then calculates 3D ground coordinates using the slope and intercept of straight lines in the EPIs, which are generated by slicing the 3D spatio‐temporal solid data along a temporal dimension. The early experimental results demonstrated that the digital elevation model (DEM) using this technique is quite rough. The improvements of mathematical model and computational procedures for increasing the DEM accuracy are presented in this paper. The improvement of the mathematical model is implemented via rectifying those possible errors without directly solving the rectification parameters, and the improvement of the computational process is implemented via avoiding nonlinear iteration of rectification process. Three test fields in Berlin, Germany were estab...

Timely and accurate change detection of Earth's surface features is extremely important for understanding relationships and interactions between human and natural phenomena in order to promote better decision making. Remote sensing data are primary sources extensively used for change detection in recent decades. Many change detection techniques have been developed. This paper summarizes and reviews these techniques. Previous literature has shown that image differencing, principal component analysis and post-classification comparison are the most common methods used for change detection. In recent years, spectral mixture analysis, artificial neural networks and integration of geographical information system and remote sensing data have become important techniques for change detection applications. Different change detection algorithms have their own merits and no single approach is optimal and applicable to all cases. In practice, different algorithms are often compared to find the best change detection results for a specific application. Research of change detection techniques is still an active topic and new techniques are needed to effectively use the increasingly diverse and complex remotely sensed data available or projected to be soon available from satellite and airborne sensors. This paper is a comprehensive exploration of all the major change detection approaches implemented as found in the literature.

Techniques based on multi-temporal, multi-spectral, satellite-sensor-acquired data have demonstrated potential as a means to detect, identify, map and monitor ecosystem changes, irrespective of their causal agents. This review paper, which summarizes the methods and the results of digital change detection in the optical/infrared domain, has as its primary objective a synthesis of the state of the art today. It approaches digital change detection from three angles. First, the different perspectives from which the variability in ecosystems and the change events have been dealt with are summarized. Change detection between pairs of images (bi-temporal) as well as between time profiles of imagery derived indicators (temporal trajectories), and, where relevant, the appropriate choices for digital imagery acquisition timing and change interval length definition, are discussed. Second, pre-processing routines either to establish a more direct linkage between remote sensing data and biophysical phenomena, or to temporally mosaic imagery and extract time profiles, are reviewed. Third, the actual change detection methods themselves are categorized in an analytical framework and critically evaluated. Ultimately, the paper highlights how some of these methodological aspects are being fine-tuned as this review is being written, and we summarize the new developments that can be expected in the near future. The review highlights the high complementarity between different change detection methods.

This paper compares segmentation results of synthetic aperture radar (SAR) images using Gaussian-Markov random field (MRF) and Gamma-MRF models. A Gamma distribution function is more accurate and proper to trace the multilook SAR intensity data distribution. However, it is found that, at least from examples used in the paper, when the distribution function is incorporated with the MRF model to implement SAR image segmentation, the Gamma-MRF model is not necessarily shown to be superior to the Gaussian-MRF model. Occasionally the Gamma-MRF model wrongly merges a few small segments, suggesting that the Gaussian-MRF model might be more stable and reliable.

Change detection from remote sensing data is often done by simple overlay of classified maps. However, such analyses can contain a significant proportion of boundary errors, especially when combining data from different sensors. This paper presents a protocol that allows reliable post-classification comparisons by taking into account classification accuracies, landscape fragmentation, planimetric accuracies, pixel sizes and grid origins. The proposed protocol has been applied, with little extra effort, in a fragmented agricultural Mediterranean zone using MSS (1970s) and TM (1990s) images. Applying the protocol, change detection had an accuracy of 85.1%, while for a direct overlay it was only 43.9% accurate. The drawback of this method is that it reduces the useful area of comparison. As the accuracy of individual classifications is critical, the paper also describes and tests a hybrid classifier that combines an unsupervised classification approach with training areas. This approach has proved more succe...

Semantic segmentation using fine-resolution remotely sensed images plays a critical role in many practical applications, such as urban planning, environmental protection, natural and anthropogenic landscape monitoring, etc. However, the automation of semantic segmentation, i.e., automatic categorization/labeling and segmentation is still a challenging task, particularly for fine-resolution images with huge spatial and spectral complexity. Addressing such a problem represents an exciting research field, which paves the way for scene-level landscape pattern analysis and decision making. In this paper, we propose an approach for automatic land segmentation based on the Feature Pyramid Network (FPN). As a classic architecture, FPN can build a feature pyramid with high-level semantics throughout. However, intrinsic defects in feature extraction and fusion hinder FPN from further aggregating more discriminative features. Hence, we propose an Attention Aggregation Module (AAM) to enhance multi-scale feature learning through attention-guided feature aggregation. Based on FPN and AAM, a novel framework named Attention Aggregation Feature Pyramid Network (A2-FPN) is developed for semantic segmentation of fine-resolution remotely sensed images. Extensive experiments conducted on three datasets demonstrate the effectiveness of our A2 -FPN in segmentation accuracy. Code is available at https://github.com/lironui/A2-FPN.

In the last 20 years, the use of fractal dimension indices has found wide application in various disciplines, including remote sensing. Recent studies have improved our understanding of both scaling and statistical properties of natural images and it is now broadly accepted that the fractal dimension, used as a sole texture descriptor, does not provide complete textural information. Natural images are characterized by a set of fractal dimensions (i.e. multifractal indices) rather than by a single fractal dimension. This paper aims to give a brief introduction to the theory of fractals and multifractals and to confirm, by means of examples, the inadequacy of using a single fractal dimension in image analysis. In addition, it will show how multifractal indices can provide better textural description. The main purpose of this paper is to prove that multifractal descriptors can substitute or complement classical textural descriptors by presenting a method to calculate the multifractal spectrum of pixels insid...

This paper aims to bring together most of the key issues involved in research in novel systems in telegeoprocessing. Telegeoprocessing can be defined as a new discipline based on real-time spatial databases updated regularly by means of telecommunications systems in order to support problem solving and decision-making at any time and any place . It involves the integration of remote sensing, Geographic Information Systems (GIS), Global Positioning System (GPS) and telecommunications. Telegeoprocessing can be categorized as WebGIS (an integration of Internet and GIS), Mobile Geoprocessing (Wireless Mapping - an integration of mobile computing and GIS) and TeleGIS (an integration of GIS and other telecommunication techniques). This paper presents a comprehensive introduction to the problems encountered in telegeoprocessing engineering and the major technologies and standards related to designing an integrated, fully functional telegeoprocessing system based on the latest multimedia and telecommunication tec...

We discuss a critical point of the paper. It is shown on the one hand that the author of this paper has mistaken the change of spectral content with the change in spatial resolution, and on the other hand, that the protocol used to establish the advantages of his smoothing filter-based intensity modulation (SFIM) technique over other methods is not appropriate at all. Though the SFIM technique has its merits, the article does not demonstrate its expected qualities.

This paper presents theoretical and practical assessments of the accuracy with which Digital Elevation Models (DEMs) can be interpolated to higher resolutions and demonstrates that simple bilinear or bicubic convolution is an adequate approach. It is assumed that the DEM is defined on a square grid and is free of error. In other words, this paper does not consider the question of how irregularly distributed elevation measurements should optimally be resampled on to a square grid. The achievable accuracy depends principally on the fractal dimension of the surface and the standard deviations of the height difference between adjacent grid points in the original DEM. The accuracy depends only slightly on the factor by which the resolution of the DEM is increased, and it is typically between 0.2 sigma and 0.6 sigma. The paper briefly discusses the implications of these results for one remote sensing application: the correction of Synthetic Aperture Radar imagery for the effects of terrain-induced distortion.

In this paper, we demonstrate that image processing algorithms (IPAs) can perform a new visual task, namely, selecting and locating regions-of-interest (ROIs). This task has been defined on the basis of a theory of human vision: the scanpath theory. Further, using the measures Sp and Ss, the similarity of location and ordering, respectively, developed over the years in studying human perception and the active looking role of eye movements, we could quantify the efficient and efficacious manner by which IPAs can imitate human vision in locating ROIs. The methodology in this paper is applied to the Mars Rover exploration.

In this paper we describe a method for recognizing and extracting the road network in peri-urban areas using SPOT panchromatic images. A particular combination of image representation description algorithms is proposed, which recognizes road features-not clearly defined in remotely sensed images and often confused with other features- and extracts them. The method consists of five algorithms-thresholding, morphological, thinning, linking, and gap filling-that arc used sequentially. The only human intervention required is the definition of a threshold. The proposed approach produces a raster road network representation that is highly complete and locationally accurate. Some experimental results are given in this paper.

This paper explores the use of generalized linear models (GLMs) for enhancing standard methods of satellite-based land-cover change detection. It starts by generalizing satellite-based change-detection algorithms in a modelling context and then gives an overview of GLMs. It goes onto describe how GLMs can fit into the context of existing change-detection methods. By way of example, using a change detection over two locations in North Carolina, USA, using Landsat Thematic Mapper data, it shows how the models provide a quantitative approach to image-based change detection. The application of GLMs requires special consideration of the spatial correlation of geographical data and how this effects the use of GLMs. The paper describes the use of preliminary variogram analysis on the image data for initial sampling considerations. For the binary response (change/no-change) derived from the reference data, a 'joint-count' test is used to assess their independence. Finally, the model error term is checked through ...

Considering that the hyperspectral image (HSI) has a large number of spectrum bands, to optimize the features and make full use of more informative features, many papers have introduced attention m...

In this paper we show some alternative classifiers to the widely used maximum likelihood (ML) classifier in order to obtain high accuracy classifications. The ML classifier does not provide high accuracy classifications when the training sets are high-overlapping in the representation space due to the shape of the decision boundaries it imposes. In these cases, it is preferred to adopt another classifier that may adjust the decision boundaries in a better fashion. This objective may be achieved with several non-parametric classifiers and by using the regularized discriminant classifier, as shown in this paper.

Over the past decade there have been considerable increases in both the quantity of remotely sensed data available and the use of neural networks. These increases have largely taken place in parallel, and it is only recently that several researchers have begun to apply neural networks to remotely sensed data. This paper introduces this special issue which is concerned specifically with the use of neural networks in remote sensing. The feed-forward back-propagation multi-layer perceptron (MLP) is the type of neural network most commonly encountered in remote sensing and is used in many of the papers in this special issue. The basic structure of the MLP algorithm is described in some detail while some other types of neural network are mentioned. The most common applications of neural networks in remote sensing are considered, particularly those concerned with the classification of land and clouds, and recent developments in these areas are described. Finally, the application of neural networks to m...

It is well known that higher dimensional information essentially leads to better accuracy in remotely sensed image classification. This paper is aimed at land cover classification from SPOT-HRV imagery by the integration of multispectral intensity and texture information. In particular, fractal dimensions are extracted using a wavelet transform as image texture. A neural network approach to classification is adopted in this paper. The underlying network is a modified multilayer perceptron trained by a Kalman filtering technique. The main advantages of this network are (1) its non-backpropagation fashion of learning which leads to a fast convergence, (2) a built-in optimization function, and (3) global scale. Saving computer storage space and a fast learning capability are in particular suitable features for remote sensing applications. Correlation analysis was subsequently performed on both the intensity and fractal images. It was found that fractal information significantly improves the discrimi...

This paper examines a number of experimental investigations of neural networks used for the classification of remotely sensed satellite imagery at the Joint Research Centre over a period of five years, and attempts to draw some conclusions about 'best practice' techniques to optimize network training and overall classification performance. The paper examines best practice in such areas as: network architecture selection; use of optimization algorithms; scaling of input data; avoidance of chaos effects; use of enhanced feature sets; and use of hybrid classifier methods. It concludes that a vast body of accumulated experience is now available, and that neural networks can be used reliably and with much confidence for routine operational requirements in remote sensing.

Data from interferometric synthetic aperture radar (INSAR) can provide three-dimensional information by using the phase as an additional information source derived from the complex radar data. In this paper, the issues, techniques and applications of SAR interferometry are reviewed. After a brief historical review, the geometric implementations of SAR interferometry are described. The general processing techniques are summarized and an introduction to various phase unwrapping techniques is given. Besides the differential use of SAR interferometry, the review focuses on parameters such as baseline, decorrelation and motion compensation which have a limiting influence on the quality of the data. After evaluating the various existing and potential applications using SAR interferometric techniques such as topographic mapping, digital elevation modelling, slope measurement, change detection, classification, ocean currents, polar research, seismic events and volcanic hazards, the paper concludes with s...

Imaging Satellites acquire multispectral images, MIs, in low resolution, LR, and panchromatic images, PANs, in high resolution, HR, due to some advantages provided in satellite design. The pansharpening, PS, is a super resolution image synthesis method that is used to generate the pansharpened image, PI, by fusion of PAN and MI. The use of PS process is unavoidable in applications such as efficient use of communication-bandwidth of imaging satellites and fusion of images derived from various image sensors. The PS process is a multi-step process consisting of various complex image processing stages, such as registration, resampling, synthesis, and fusion. The Weighted Differential Evolution Algorithm-based PS method, WDEPS, has been proposed in this paper. The WDEPS uses WDE to synthesize the intensity image, which is the blended-image of MI -bands. WDE has been used to compute the relevant image-blending weights, efficiently. In the experiments, several satellite images (QuickBird-2, Ikonos-2, and GeoEye-1) with different spatial resolutions were used. The WDEPS's experimental results have been compared with 17 well-known PS methods by using 9 full-reference and 3 blind image quality assessment metrics. The experimental results exposed that WDEPS generates high-quality PIs than traditional PS methods used in the experimental aspect of qualitative and quantitative assessment.

Feature matching in optical remote sensing images usually suffers from significant intensity differences and background variations. In this paper, an effective image matching approach is proposed t...

Research on spatial information is a hot topic in hyperspectral remote-sensing image (HSI) exploitation. This paper defines a spatial correlation filter to study the spatial variation characteristi...

Over the past decade the Argos System has become a widely used tool for animal tracking. This paper investigates whether lower costs and energy requirements could be achieved by alternative techniques. These include the global radionavigation systems GPS and Omega which employ low-powered receivers rather than the high-powered transmitters of Argos. A variety of systems are studied and their costs and energy requirements modelled within specific marine mammal tracking scenarios. This paper presents a fundamental analysis exploring current and future options available to researchers who need to track marine mammals.

Conventional classification techniques, both supervised and non-supervised, are used for the classification of remote sensing data on the basis of spectral signatures of the classes of interest, and, contextual techniques use both spectral and spatial information to improve the classification accuracies. This paper reviews some of the recent contextual classification techniques and proposes two methods. One method is meant for low resolution and the other for high resolution images. Test results for both the methods of two sets of data, corresponding to low resolution and high resolution, respectively, are compared with Gaussian Maximum Likelihood (GML) results and presented in the paper.

This paper presents an overview of the development of the Earth Observation Data Centre (EODC) and its facilities for processing and archiving ERS-1 data. The hardware systems of the EODC and the implementation of product processing chains are described. The paper concludes with a look at the achievements of the EODC and a discussion of the problems encountered with a view to the lessons learnt for future developments.

In the previous paper we derived the fundamental noise limits in the determination of the autofocus parameter in SAR processing for both point and extended objects. In the present paper we extend this analysis to establish the fundamental limits on the image quality of a phase-corrected SAR image. A variety of autofocus schemes are analysed and an improved scheme is proposed which offers the minimum long-term image distortion and reasonable point response to changes in the autofocus parameter across the full aperture. The implications for phase correction based on autofocusing, leading to distortion-free images are examined. Image distortion, phase error and the point-target response are all derived as a function of the extent to which the full aperture is subdivided so as to match high-frequency changes in the focus parameter.

This paper discusses basic mathematical formulations to clarify the problem of automatic segmentation of sea-ice remote sensing imagery. The work is illustrated with a SAR image. The main issue is a robust and automatic analysis of sea-ice conditions from remotely sensed data. The paper reveals some of the potential and essential features that are hidden in a class of filters often known as simple ‘shrink and expand operators’. The intensity of radiation from the ground is idealized as a measure which may have singularities. Absolutely calibrated digital images represent this measure where the measurable set are restricted to unions of picture elements. A related proposed martingale stopping time scheme makes image segmentation robust to changes in resolution. Tunable segmentations based on mathematical morphology can be controlled to give optimal ‘physically proper’ sets which are independent of calibration. This approach introduces physical knowledge, time aspects, and random geometry to the pr...