for Journals by Title or ISSN
for Articles by Keywords
help
Followed Journals
Journal you Follow: 0
 
Sign Up to follow journals, search in your chosen journals and, optionally, receive Email Alerts when new issues of your Followed Journals are published.
Already have an account? Sign In to see the journals you follow.
Similar Journals
Journal Cover
Selected Topics in Applied Earth Observations and Remote Sensing, IEEE Journal of
Journal Prestige (SJR): 1.547
Citation Impact (citeScore): 4
Number of Followers: 53  
 
  Hybrid Journal Hybrid journal (It can contain Open Access articles)
ISSN (Print) 1939-1404
Published by IEEE Homepage  [191 journals]
  • Frontcover
    • Abstract: Presents the front cover for this issue of the publication.
      PubDate: April 2019
      Issue No: Vol. 12, No. 4 (2019)
       
  • IEEE Geoscience and Remote Sensing Societys
    • Abstract: Presents a listing of the editorial board, board of governors, current staff, committee members, and/or society editors for this issue of the publication.
      PubDate: April 2019
      Issue No: Vol. 12, No. 4 (2019)
       
  • IEEE Geoscience and Remote Sensing Societys
    • Abstract: Presents a listing of the editorial board, board of governors, current staff, committee members, and/or society editors for this issue of the publication.
      PubDate: April 2019
      Issue No: Vol. 12, No. 4 (2019)
       
  • Institutional listings
    • Abstract: The IEEE Geoscience and Remote Sensing Society provides paid Institutional Listings from firms interested in the field of geoscience and remote sensing.
      PubDate: April 2019
      Issue No: Vol. 12, No. 4 (2019)
       
  • Estimation of Significant Wave Height of Near-Range Traveling Ocean Waves
           Using Sentinel-1 SAR Images
    • Authors: Fabian Surya Pramudya;Jiayi Pan;Adam Thomas Devlin;
      Pages: 1067 - 1075
      Abstract: Synthetic aperture radar (SAR) is a valuable tool to observe many oceanographic parameters of the ocean surface. The significant wave height (Hs or SWH), defined as the mean wave height of the highest one-third of waves, can be estimated using indirect relations between SAR-derived and the actual ocean wave spectra. However, multilook SAR imaging of ocean surface waves is highly affected by degraded azimuthal resolution and the random shift of scattered elements as the waves propagate off the SAR range direction. Therefore, the azimuthal displacement is found to be very low when near-range travelling ocean waves with narrow band spectrum are imaged. In this study, an improved algorithm is developed to directly estimate the significant wave height of ocean waves that propagate in the near-range direction using SARSAR imagery. The slope of the NRCS with respect to the radar wave incidence angle is derived from SAR images using an iterative scheme. A total of 69 SAR images from Sentinel-1A and 1B from 2016 to 2017, acquired near Hawaii, and in situ wave data from nearby National Data Buoy Center buoys are used to estimate Hs and validate the improved algorithm. Results show that the algorithm performs well in estimating SWH under low to moderate wind forcing conditions (4–10 ms−1). The accuracy of the SWH estimation decreases under high-wind-speed and wind-wave-dominant conditions.
      PubDate: April 2019
      Issue No: Vol. 12, No. 4 (2019)
       
  • A New Methodology for Rice Area Monitoring With COSMO-SkyMed HH–VV
           PingPong Mode SAR Data
    • Authors: Lucio Mascolo;Giuseppina Forino;Ferdinando Nunziata;Giovanni Pugliano;Maurizio Migliaccio;
      Pages: 1076 - 1084
      Abstract: In this paper, a novel approach is proposed to exploit a time series of COSMO-SkyMed (CSK) HH-VV SAR images to map rice fields and to estimate the sowing dates. The approach relies on multi-polarization features, i.e., the squared modulus of the HH and VV channels and the polarization ratio, extracted from CSK SAR scenes. The key step consists of extracting a rice training signature related to the multipolarization features. This signature allows estimating the sowing date that, at once, is used to refine the rice map obtained by the conventional interpretation of the CSK time series in terms of the scattering mechanisms of the different growing cycles. Experiments, carried out on a time series of 32 CSK images, collected from the Mekong Delta region, South Vietnam, confirm the soundness of the proposed methodology which is shown to provide results comparable to the ones obtained by a literature approach that exploits a similar dataset.
      PubDate: April 2019
      Issue No: Vol. 12, No. 4 (2019)
       
  • Responses of Vegetation Productivity to Temperature Trends Over
           Continental United States From MODIS Imagery
    • Authors: Peng Fu;
      Pages: 1085 - 1090
      Abstract: Vegetation growth and productivity may record signals of global climate change since vegetation is sensitive to modifications in climatic variables, such as temperature and precipitation. Significant changes in vegetation growth and productivity can affect the exchange of energy, water vapor, and momentum between the land surface and atmosphere. Despite consensus on the vegetation growth enhancement by the warming of temperature, little attention has been given to the understanding of the impacts of cooling temperature trends on vegetation growth and productivity. Based on satellite-derived land surface temperatures and vegetation indices, this study shows that alternating temperature trends are more reasonable than a monotonic temperature trend for the study period from 2003 to 2016. More importantly, results reveal that both warming and cooling temperature trends can have a positive impact on vegetation growth over the continental United States. These findings are of importance to understand global carbon cycle sinks and sources accurately in different regions.
      PubDate: April 2019
      Issue No: Vol. 12, No. 4 (2019)
       
  • Geo-Object-Based Soil Organic Matter Mapping Using Machine Learning
           Algorithms With Multi-Source Geo-Spatial Data
    • Authors: Tianjun Wu;Jiancheng Luo;Wen Dong;Yingwei Sun;Liegang Xia;Xuejian Zhang;
      Pages: 1091 - 1106
      Abstract: Soil is a complicated historical natural continuum that presents gradual changes in its properties and geographic area. Conventional soil survey and cartography methods on a macroscopic scale based on grids with a coarse resolution are inadequate for the rapid development of precision agriculture. The demand for soil mapping content and accuracy has increased as more convenient methods of acquiring multi-source geo-spatial data have been developed, and such data are commonly employed to extract basic mapping units and environmental variables in related algorithms. We employ geo-objects as basic units of soil property mapping, which are extracted from high-resolution remote sensing images using a convolutional neural network based learning algorithm. Multi-source geo-spatial data are transferred into each geo-object as environmental variables, and the relationships between soil properties and environmental variables are mined using powerful tree-based machine learning algorithms, including regressions with random forests and XGBoost. A data set that includes soil sample points and multi-source geo-spatial data is used to evaluate the effectiveness of the proposed method. The experimental results demonstrate that the method allows for better soil organic matter mapping than state-of-the-art interpolation-based and linear-regression-based methods. The proposed procedure has potential to be a general method for mapping other soil properties. Its advantages are embodied in the modeling of relatively miscellaneous data with implicitly associated non-linear relationships between soil properties and environmental variables. The spatial scale and accuracy of the finer maps capture more detailed characteristics of the soil properties and are applicable to the micro-domain fields required for refined soil mapping with small variations.
      PubDate: April 2019
      Issue No: Vol. 12, No. 4 (2019)
       
  • Microwave and Meteorological Fusion: A method of Spatial Downscaling of
           Remotely Sensed Soil Moisture
    • Authors: Hao Sun;Chuangchuang Cai;Hongxing Liu;Bo Yang;
      Pages: 1107 - 1119
      Abstract: Downscaling of microwave remotely sensed that soil moisture content (SMC) is an efficient way to obtain spatial continuous SMC at a finer resolution. However, the classical optical/thermal and microwave fusion, and the active and passive microwave fusion cannot work under all-weather conditions because of contamination of clouds or the lack of suitable radar data source. In this study, a microwave and meteorological fusion (MMF) is provided. The MMF method is based on a complementary relationship hypothesis assuming SMC is reflected in the adjacent surface atmospheric moisture under midday conditions. By this method, daily passive SMC products from Soil Moisture Active Passive (SMAP) mission with 36-km resolution were disaggregated using a daily gridded meteorological data with nominal 4-km resolution. The original and downscaled SMCs were evaluated by comparing with in situ SMC obtained from three core validation sites and three sparse networks. The experiment was conducted in the central part of the U.S. from April 2015 to June 2018. Results demonstrated that the downscaled SMC maintained the dynamic range of original SMC product and energy was conserved. Furthermore, the downscaled SMC showed good agreement with and slightly outperformed the original SMC as compared with in situ SMC. The downscaling method is shown to capture higher resolution SMC spatial variability while preserving the quality of original SMC. However, because of the complexity of soil moisture–atmosphere interactions, the actual contributing domain of downscaled SMC may be greater than 4 km. The MMF method is suggested as a supplementary for all-weather downscaling coarse-resolution SMC.
      PubDate: April 2019
      Issue No: Vol. 12, No. 4 (2019)
       
  • Integration of Satellite Images and Open Data for Impervious Surface
           Classification
    • Authors: Zelang Miao;Yuelong Xiao;Wenzhong Shi;Yueguang He;Paolo Gamba;Zhongbin Li;Alim Samat;Lixin Wu;Jia Li;Hao Wu;
      Pages: 1120 - 1133
      Abstract: Supervised learning is vital to classify impervious surface from satellite images. Despite its effectiveness, the training samples need to be provided manually, which is time consuming and labor intensive, or even impractical when classifying satellite images at the regional/global scale. This study, therefore, sets out to automatically generate training samples from open data, based on the fact that cities and urban areas are nowadays full of individual geo-referenced data, such as social network data. The proposed method consists of automatic generation of training samples based on a filtering process of open data, satellite image pre-processing, and impervious surface detection using one class classification (OCC). Two Landsat-8 Operational Land Imager images were selected to test the proposed method. The results show that the proposed method is effective in impervious surface with good classification accuracy. The findings in this study shine new light on the applications of open data in remote sensing.
      PubDate: April 2019
      Issue No: Vol. 12, No. 4 (2019)
       
  • Intra-Urban Change Analysis Using Sentinel-1 and Nighttime Light Data
    • Authors: Meiqin Che;Paolo Gamba;
      Pages: 1134 - 1142
      Abstract: This paper is devoted to detect and classify intra-urban changes by jointly exploiting Sentinel-1 (S-1) SAR data and nighttime light data. By extracting urban extents and urban density maps from SAR data, changes in nighttime lights can be used to detect changes related to the level of activity in a specific portion of each urban areas. At the same time, changes in radar backscattering are prone to reveal changes in the two- and three-dimensional structures of the built-up. The combination of these multimodal datasets has already proved to be useful to discriminate urban change patterns at the city level. In this paper, instead, SAR datasets from S-1 are exploited, allowing the recognition of different intra-urban changes. Experimental results focus on fast growing (mega) cities in East Asia, allowing us to understand in a more detailed way how they are changing and evolving in all three dimensions. Examples for Nanjing, Shanghai, and Guangzhou (China), Saigon (Vietnam), and Vientiane (Laos) are discussed to prove this statement.
      PubDate: April 2019
      Issue No: Vol. 12, No. 4 (2019)
       
  • Mapping Global Urban Areas From 2000 to 2012 Using Time-Series Nighttime
           Light Data and MODIS Products
    • Authors: Zuoqi Chen;Bailang Yu;Yuyu Zhou;Hongxing Liu;Chengshu Yang;Kaifang Shi;Jianping Wu;
      Pages: 1143 - 1153
      Abstract: Mapping urban dynamics at the global scale becomes a pressing task with the increasing pace of urbanization and its important environmental and ecological impacts. In this study, we proposed a new approach to mapping global urban areas from 2000 to 2012 by applying a region-growing support vector machine classifier and a bidirectional Markov random field model to time-series nighttime light data. In this approach, both spectrum and spatial-temporal contextual information are employed for an improved urban area mapping. Our results indicate that at the global level, the urban area increased from 625,000 to 1,039,000 km2 during 2000–2012. Most urban areas are concentrated in the region between 30°N and 60°N latitudes. The latitudinal distribution of urban areas from this study is consistent with three land-cover products, including European Space Agency Climate Change Initiative Land Cover dataset, Finer Resolution Observation and Monitoring Global Land Cover, and 30-m Global Land Cover dataset. We found that for several major cities, such as Shanghai, urban areas from our study contain some nonurban land-cover types with intensive human activities. The validation using Landsat 7 ETM+ imagery indicates that the overall accuracies of the mapped urban areas for 2000, 2005, 2008, and 2010 are 86.0%, 88.6%, 89.8%, and 88.7%, respectively, and the Kappa coefficients are 0.72, 0.77, 0.79, and 0.78, respectively. This study also demonstrates that the integration of the spatial-temporal contextual information and the use of bidirectional Markov random field model are effective in improving the accuracy and temporal consistency of urban area mapping using time-series nighttime light data.
      PubDate: April 2019
      Issue No: Vol. 12, No. 4 (2019)
       
  • Robust Human Targets Tracking for MIMO Through-Wall Radar via
           Multi-Algorithm Fusion
    • Authors: Huquan Li;Guolong Cui;Lingjiang Kong;Guohao Chen;Mingyang Wang;Shisheng Guo;
      Pages: 1154 - 1164
      Abstract: The detection and tracking of human targets behind the wall is of great importance in urban sensing. The random and high maneuvering behaviors of moving human targets and the clutter diversity lead to high missed detection and false-alarm probability. In this paper, we consider two-dimensional human-target tracking problem, and a multi-algorithm fusion (MAF) framework exploiting the mean-shift algorithm and the Kalman filter is proposed. Compared with the mean-shift algorithm, the proposed MAF framework has robust tracking performance, especially in the presence of multiple targets. Finally, the proposed MAF framework is evaluated by simulations and real data.
      PubDate: April 2019
      Issue No: Vol. 12, No. 4 (2019)
       
  • Errors in Time-Series Remote Sensing and an Open Access Application for
           
    • Authors: Brad G. Peter;Joseph P. Messina;
      Pages: 1165 - 1174
      Abstract: Remotely sensed measures of productivity are frequently used to characterize global agriculture and vegetated ecosystems, and are often downscaled to describe local, remote areas where finer spatial and temporal resolution data are regularly unavailable. While data errors may propagate throughout any analytical procedure, those that are missed during delivery and preliminary data mining require more attention. Here, a collection of formerly and presently available global remote sensing products are compiled to demonstrate the temporal and geographic breadth of remote sensing uncertainty. Vegetation productivity measures are invaluable for monitoring global health, but erroneous estimates that go unrecognized may result in serious policy mistakes. It is eminently clear that generalizable and accessible a priori methods for anomaly detection are lacking and urgently needed so that data errors are recognized before public delivery and before widespread use. Simple yet effective statistics such as the modified Z-score, Tukey's outliers, and Geary's C are leveraged here to identify, locate, and visualize the types of outliers that remote sensing data users may elect to omit or correct. Contributing to the growing ensemble of Google Earth Engine methodologies, we propose this generalizable method of detecting spatial outliers for remote sensing error management by users across scientific domains.
      PubDate: April 2019
      Issue No: Vol. 12, No. 4 (2019)
       
  • Ground Feature Oriented Path Planning for Unmanned Aerial Vehicle Mapping
    • Authors: Chun Liu;Shuhang Zhang;Akram Akbar;
      Pages: 1175 - 1187
      Abstract: Unmanned aerial vehicles (UAVs) are being used to take roles that were previously performed by traditional manned aircraft, such as remote sensing and photogrammetry. The standard path planning for UAV mapping is mainly executed by adopting the “lawnmower” mode. However, some situations that have sparse or repetitive features are problematic to map with this technique, given that orthoimage stitching relies heavily on the number and quality of image tie points. Traditional path planning can result in some unregistered images due to a lack of tie points. This paper proposes a ground feature oriented path-planning method for UAV mapping. The method first estimates the distribution of the ground feature points from a lower-resolution image. Then, image footprints are selected by applying a three-step optimization. The flight path for the UAV is then generated by solving the “grouped traveling salesman” problem. This approach ensures the georegistration of images during orthoimage stitching while maximizing the orthoimage coverage. Two cases, including a simulation and a real-world case, together with standard path-planning modes with different overlaps, are selected to evaluate the proposed method. The results show that the proposed method covers the same area with the smallest number of images. The model excludes problematic areas from the scanning path to generate a more efficient processing dataset.
      PubDate: April 2019
      Issue No: Vol. 12, No. 4 (2019)
       
  • Pansharpening via Detail Injection Based Convolutional Neural Networks
    • Authors: Lin He;Yizhou Rao;Jun Li;Jocelyn Chanussot;Antonio Plaza;Jiawei Zhu;Bo Li;
      Pages: 1188 - 1204
      Abstract: Pansharpening aims to fuse a multispectral (MS) image with an associated panchromatic (PAN) image, producing a composite image with the spectral resolution of the former and the spatial resolution of the latter. Traditional pansharpening methods can be ascribed to a unified detail injection context, which views the injected MS details as the integration of PAN details and bandwise injection gains. In this paper, we design a new detail injection based convolutional neural network (DiCNN) framework for pansharpening with the MS details being directly formulated in end-to-end manners, where the first detail injection based CNN (DiCNN1) mines MS details through the PAN image and the MS image, and the second one (DiCNN2) utilizes only the PAN image. The main advantage of the proposed DiCNNs is that they provide explicit physical interpretations and can achieve fast convergence while achieving high pansharpening quality. Furthermore, the effectiveness of the proposed approaches is also analyzed from a relatively theoretical point of view. Our methods are evaluated via experiments on real MS image datasets, achieving excellent performance when compared to other state-of-the-art methods.
      PubDate: April 2019
      Issue No: Vol. 12, No. 4 (2019)
       
  • Fusion of Hyperspectral and Multispectral Images Based on a Bayesian
           Nonparametric Approach
    • Authors: Lichun Sui;Li Li;Jonathan Li;Nan Chen;Yongqing Jiao;
      Pages: 1205 - 1218
      Abstract: This paper presents a new approach to fusion of hyperspectral and multispectral images based on Bayesian nonparametric sparse representation. The approach formulates the image fusion problem within a constrained optimization framework, while assuming that the target image lives in a lower dimensional subspace. The subspace transform matrix is determined by principal component analysis, and the sparse regularization term is designed depending on a set of dictionaries and sparse coefficients associated with the observed images. Specifically, the dictionary elements and sparse coefficients are learned by the Bayesian nonparametric approach with the beta-Bernoulli process, which establishes the probability distribution models for each latent variable and calculates the posterior distributions by Gibbs sampling. Finally, serving the obtained posterior distributions as a priori, the fusion problem is solved via an alternate optimization process, where the alternate direction method of multipliers is applied to perform the optimization with respect to the target image. The Bayesian nonparametric method is used to optimize the sparse coefficients. Exhaustive experiments using both two public datasets and one real-world dataset of remote sensing images show that the proposed approach outperforms the existing state-of-the-art methods.
      PubDate: April 2019
      Issue No: Vol. 12, No. 4 (2019)
       
  • Quantitative Reconstruction and Denoising Method HyBER for Hyperspectral
           Image Data and Its Application to CRISM
    • Authors: Linyun He;Joseph A. O’Sullivan;Daniel V. Politte;Kathryn E. Powell;Raymond E. Arvidson;
      Pages: 1219 - 1230
      Abstract: We propose a new algorithm, hypothesis-based estimation with regularization (HyBER), to reconstruct and denoise hyperspectral image data without extra statistical assumptions. The hypothesis test selects the best statistical model approximating measurements based on the data only. A regularized maximum log-likelihood estimation method is derived based on the selected model. A spatially dependent weighting on the regularization penalty is presented, substantially eliminating row artifacts that are due to nonuniform sampling. A new spectral weighting penalty is introduced to suppress varying detector-related noise. HyBER generates reconstructions with sharpened images and spectra in which the noise is suppressed, whereas fine-scale mineral absorptions are preserved. The performance is quantitatively analyzed for simulations with 0.002% relative error, which is better than the traditional nonstatistical methods (baselines) and statistical methods with improper assumptions. When applied to the Mars Reconnaissance Orbiter's Compact Reconnaissance Imaging Spectrometer for Mars data, the spatial resolution and contrast are approximately two times better as compared to map projecting data without the use of HyBER.
      PubDate: April 2019
      Issue No: Vol. 12, No. 4 (2019)
       
  • Noniterative Hyperspectral Image Reconstruction From Compressive Fused
           Measurements
    • Authors: Jorge Bacca;Claudia V. Correa;Henry Arguello;
      Pages: 1231 - 1239
      Abstract: Compressive spectral imaging (CSI) enables the acquisition of spectral and spatial information of a scene using fewer projected measurements than traditional scanning approaches. Recently, research efforts have focused on obtaining high-resolution spectral images via expensive detectors and sophisticated CSI devices. Alternatively, high-resolution spectral images can be obtained using side information or fusion of compressed measurements, without significantly increasing acquisition costs. Indeed, these approaches retrieve improved resolution images applying iterative and computationally expensive algorithms. This paper proposes the fusion of compressed measurements obtained from two state-of-the-art CSI systems, the single-pixel camera (SPC) and the three-dimensional coded aperture snapshot imaging system (3D-CASSI), such that high-resolution images can be obtained by exploiting detailed spectra provided by the SPC and high spatial resolution of the 3D-CASSI. Specifically, a noniterative reconstruction algorithm is proposed, based on the fact that the spatial–spectral data lie in a low-dimensional subspace. In contrast to related works, the proposed approach relies on implementable CSI systems. Simulations and experimental results show the effectiveness of the proposed method compared to similar approaches, both in reconstruction quality and complexity. Specifically, the proposed method is up to 5.6 times faster than its counterparts and provides comparable quality of attained reconstructions.
      PubDate: April 2019
      Issue No: Vol. 12, No. 4 (2019)
       
  • An Efficient Residual Learning Neural Network for Hyperspectral Image
           Superresolution
    • Authors: Wei Liu;Joonwhoan Lee;
      Pages: 1240 - 1253
      Abstract: Deep learning, especially a discriminative model for image reconstruction, has shown great potential for single image superresolution (SR) of hyperspectral images (HSI). For HSI SR task, it is crucial to predicting each pixel according to the surrounding context, exploiting both spatial and spectral correlation information simultaneously. In this paper, an efficient three-dimensional (3-D) HSI SR convolution neural network (CNN) based on residual learning is proposed. The network builds convolutional layers in low-resolution (LR) space and extracts the features along both spatial and spectral dimensions using 3-D dilated kernel. Then, 3-D deconvolution is employed at the last layer, which enlarges the image to the desired size. By employing multibranch and multiscale fusion in the architecture, the network can learn a better and more complex LR to high-resolution mapping. The overall network combines the global with local residual learning to reduce training difficulty and improve the performance. The design philosophy of our model is to find the best tradeoff between performance and computational cost. We train the network in an end-to-end fashion, and the experimental results of the quantitative and qualitative evaluation show that our proposed method yields satisfactory SR performance.
      PubDate: April 2019
      Issue No: Vol. 12, No. 4 (2019)
       
  • Target Dictionary Construction-Based Sparse Representation Hyperspectral
           Target Detection Methods
    • Authors: Dehui Zhu;Bo Du;Liangpei Zhang;
      Pages: 1254 - 1264
      Abstract: Hyperspectral imagery (HSI) with a high spectral resolution contains hundreds and even thousands of spectral bands, and conveys abundant spectral information, which provides a unique advantage for target detection. A number of classical target detectors have been proposed based on the linear mixing model (LMM) and sparsity-based model. Compared with the LMM, sparsity-based detectors present a better performance on dealing with the spectral variability. Despite the great success of the sparsity-based model in recent years, one problem with all state-of-the-art sparsity-based models still exist: the target dictionary is formed via the target training samples that are selected from the global image scene. This is an improper way to construct target dictionary for hyperspectral target detection since the priori information is usually a given target spectrum obtained from a spectral library. Besides, target training samples selected from the global image scene are usually insufficient, which results in the problem that the target training samples and background training samples are unbalanced in the data volume, causing a deteriorated detection model. To tackle these problems, this paper constructs a target dictionary construction-based method, then proposes the constructed target dictionary-based sparsity-based target detection model and the constructed target dictionary-based sparse representation-based binary hypothesis model, which are called TDC-STD and TDC-SRBBH, respectively. Both of the proposed algorithms only need a given target spectrum as the input priori information. By using the given target spectrum for pre-detection via constrained energy minimization, we choose the pixels that have large output values as target training samples to construct the target dictionary. The proposed algorithms were tested on three benchmark HSI datasets and the experimental results show that the proposed algorithms demonstrate outstanding detection performances when compared w-th other state-of-the-art detectors.
      PubDate: April 2019
      Issue No: Vol. 12, No. 4 (2019)
       
  • Abundance-Indicated Subspace for Hyperspectral Classification With Limited
           Training Samples
    • Authors: Shuyuan Xu;Jun Li;Mahdi Khodadadzadeh;Andrea Marinoni;Paolo Gamba;Bo Li;
      Pages: 1265 - 1278
      Abstract: The imbalance between the (often limited) number of available training samples and the high data dimensionality, together with the presence of mixed pixels, often complicates the classification of remotely sensed hyperspectral data. In this paper, we tackle these problems by developing a new method that combines spectral unmixing and classification techniques in a subspace-based approach. The proposed method is developed under the assumption that the spectral signature of a land cover class is associated with a given set of pure spectral signatures (called endmembers in spectral unmixing terminology), which define a low-dimensional subspace with clear physical meaning. We aim to exploit this relationship to learn the class-dependent subspaces and integrate them with a multinomial logistic regression procedure. Experiments on synthetic datasets and real hyperspectral images show that our method is able to obtain competitive performances in comparison with other approaches, particularly when very limited training sets are available.
      PubDate: April 2019
      Issue No: Vol. 12, No. 4 (2019)
       
  • Sparsity-Constrained Distributed Unmixing of Hyperspectral Data
    • Authors: Sara Khoshsokhan;Roozbeh Rajabi;Hadi Zayyani;
      Pages: 1279 - 1288
      Abstract: Spectral unmixing (SU) is a technique to characterize mixed pixels in hyperspectral images measured by remote sensors. Most of the spectral unmixing algorithms are developed using the linear mixing models. To estimate endmembers and fractional abundance matrices in a blind problem, nonnegative matrix factorization (NMF) and its developments are widely used in the SU problem. One of the constraints which was added to NMF is sparsity, that was regularized by $ L_ {q}$ norm. In this paper, a new algorithm based on distributed optimization is suggested for spectral unmixing. In the proposed algorithm, a network including single-node clusters is employed. Each pixel in the hyperspectral images is considered as a node in this network. The sparsity-constrained distributed unmixing is optimized with diffusion least mean p-power (LMP) strategy, and then the update equations for fractional abundance and signature matrices are obtained. Afterward, the proposed algorithm is analyzed for different values of LMP power and $ L_ {q}$ norms. Simulation results based on defined performance metrics illustrate the advantage of the proposed algorithm in spectral unmixing of hyperspectral data compared with other methods.
      PubDate: April 2019
      Issue No: Vol. 12, No. 4 (2019)
       
  • Semi-Supervised Hyperspectral Band Selection Based on Dynamic Classifier
           Selection
    • Authors: Xianghai Cao;Cuicui Wei;Yiming Ge;Jie Feng;Jing Zhao;Licheng Jiao;
      Pages: 1289 - 1298
      Abstract: The abundant spectral information of hyperspectral imagery makes it suitable for the classification of land cover types. However, the high dimensionality also brings some negative effects for the classification tasks. Dynamic classifier selection, in which the base classifiers are selected according to each new sample to be classified, can select the best classifier for each query sample. In this paper, a semi-supervised wrapper band selection method—the band selection based on dynamic classifier selection—is introduced to select the most discriminating bands. In the proposed method, band selection is conducted based on the selection of base classifier. Specifically, the support vector machine classification map is filtered to provide a high-quality reference, and K-nearest neighbors method is used to define the local region, finally, the band with the best classification performance is selected. Three widely used real hyperspectral datasets are used to illustrate the effectiveness of the proposed method, experimental results show that the proposed method obtains state-of-the-art performance.
      PubDate: April 2019
      Issue No: Vol. 12, No. 4 (2019)
       
  • CNN-Based Multilayer Spatial–Spectral Feature Fusion and Sample
           Augmentation With Local and Nonlocal Constraints for Hyperspectral Image
           Classification
    • Authors: Jie Feng;Jiantong Chen;Liguo Liu;Xianghai Cao;Xiangrong Zhang;Licheng Jiao;Tao Yu;
      Pages: 1299 - 1313
      Abstract: The extraction of joint spatial–spectral features has been proved to improve the classification performance of hyperspectral images (HSIs). Recently, utilizing convolutional neural networks (CNNs) to learn joint spatial–spectral features has become of great interest. However, the existing CNN models ignore complementary spatial–spectral information among the shallow and deep layers. Moreover, insufficient training samples in HSIs afflict these CNN models with overfitting problem. In order to address these problems, a novel CNN method for HSI classification is proposed. It considers multilayer spatial–spectral feature fusion and sample augmentation with local and nonlocal constraints, which is abbreviated as MSLN-CNN. In MSLN-CNN, a triple-architecture CNN is constructed to extract spatial–spectral features by cascading spectral features to dual-scale spatial features from shallow to deep layers. Then, multilayer spatial–spectral features are fused to learn complementary information among the shallow layers with detailed information and the deep layers with semantic information. Finally, the multilayer spatial–spectral feature fusion and classification are integrated into a unified network, and MSLN-CNN can be optimized in the end-to-end way. To alleviate the small sample size problem, the unlabeled samples having high confidences on local spatial constraint and nonlocal spectral constraint are selected and prelabeled. The nonlocal spectral constraint considers the structure information with spectrally similar samples in the nonlocal searching, while the local spatial constraint utilizes the contextual information with spatially adjacent samples. Experimental results on several hyperspectral datasets demonstrate that the proposed method achieves more encouraging classification performance than the current state-of-the-art classificatio- methods, especially with the limited training samples.
      PubDate: April 2019
      Issue No: Vol. 12, No. 4 (2019)
       
  • A Comparative Land-Cover Classification Feature Study of Learning
           Algorithms: DBM, PCA, and RF Using Multispectral LiDAR Data
    • Authors: Suoyan Pan;Haiyan Guan;Yongtao Yu;Jonathan Li;Daifeng Peng;
      Pages: 1314 - 1326
      Abstract: Multispectral LiDAR, characterization of completeness, and consistency of spectrum and spatial geometric data provide a new data source for land cover classification. However, how to choose the optimal features for a given set of land covers is an open problem for effective land cover classification. To address this problem, we propose a comparative scheme, which investigates a popular deep learning (deep Boltzmann machine, DBM) model for high-level feature representation and widely used machine learning methods for low-level feature extraction and selection [principal component analysis (PCA) and random forest (RF)] in land cover classification. The comparative study was conducted on the multispectral LiDAR point clouds, acquired by a Teledyne Optech's Titan airborne system. The deep learning-based high-level feature representation experimental results showed that, on an ordinary personal computer or workstation, this method required larger training samples and more computational complexity than the machine learning-based low-level feature extraction and selection methods. However, our comparative experiments demonstrated that the classification accuracies of the DBM-based method were higher than those of the RF-based and PCA-based methods using multispectral LiDAR data.
      PubDate: April 2019
      Issue No: Vol. 12, No. 4 (2019)
       
  • Waveform Inversion-Assisted Distributed Reverse Time Migration for
           Microseismic Location
    • Authors: Fangyu Li;Yan Qin;WenZhan Song;
      Pages: 1327 - 1332
      Abstract: We present a novel approach to locate microseismic sources in situ and in real time in distributed sensor networks (DSN). We propose a distributed reverse time migration (RTM) microseismic source location algorithm. RTM-based methods have advantages in passive source location in terms of robustness and accuracy. However, the traditional methods have a centralized data collection and ex situ postprocessing style, and were not designed for in situ and real-time seismic imaging, so communication and computation costs were not considered. Utilizing newly emerging DSN, real-time and in situ microseismic source location becomes possible. Thus, we specially design a joint imaging condition for the DSN system to reduce both computation and communication burdens. The tradeoff, however, is location resolution reduction. Sequentially, we employ a waveform inversion approach to obtain a finer resolution microseismic source location result. Finally, we validate the proposed method using both synthetic and field seismic records. Our approach shows promising performances using synthetics with single and multiple sources, as well as with field data. The proposed waveform inversion-assisted distributed RTM location algorithm obtains high-resolution source location results and significantly reduced communication and computation costs, even at a low signal-to-noise ratio.
      PubDate: April 2019
      Issue No: Vol. 12, No. 4 (2019)
       
  • Corrections to “Using Land Long Term Data Records to Map Land Cover
           Changes in China over 1981--2010” [Apr 17 1372-1389]
    • Authors: H. Li;P. Xiao;X. Feng;Y. Yang;L. Wang;W. Zhang;X. Wang;W. Feng;X. Chang;
      Pages: 1333 - 1334
      Abstract: Presents corrections to maps for the paper, [“Using land long term data records to map land cover changes in China over 1981–2010,” (Li, J. et al), IEEE J. Sel. Topics Appl. Earth Observ. Remote Sens., vol. 10, no. 4, pp. 1372–1389, Apr. 2017.
      PubDate: April 2019
      Issue No: Vol. 12, No. 4 (2019)
       
  • Become a published author in 4 to 6 weeks
    • Pages: 1335 - 1335
      Abstract: Advertisement, IEEE.
      PubDate: April 2019
      Issue No: Vol. 12, No. 4 (2019)
       
  • Introducing IEEE collabratec
    • Pages: 1336 - 1336
      Abstract: Advertisement, IEEE.
      PubDate: April 2019
      Issue No: Vol. 12, No. 4 (2019)
       
 
 
JournalTOCs
School of Mathematical and Computer Sciences
Heriot-Watt University
Edinburgh, EH14 4AS, UK
Email: journaltocs@hw.ac.uk
Tel: +00 44 (0)131 4513762
Fax: +00 44 (0)131 4513327
 
Home (Search)
Subjects A-Z
Publishers A-Z
Customise
APIs
Your IP address: 18.234.111.56
 
About JournalTOCs
API
Help
News (blog, publications)
JournalTOCs on Twitter   JournalTOCs on Facebook

JournalTOCs © 2009-