Probability Distributions.- Linear Models for Regression.- Linear Models for Classification.- Neural Networks.- Kernel Methods.- Sparse Kernel Machines.- Graphical Models.- Mixture Models and EM.- Approximate Inference.- Sampling Methods.- Continuous Latent Variables.- Sequential Data.- Combining Models.

Pattern Recognition and Machine Learning

Shifts in the timing of spring phenology are a central feature of global change research. Long-term observations of plant phenology have been used to track vegetation responses to climate variability but are often limited to particular species and locations and may not represent synoptic patterns. Satellite remote sensing is instead used for continental to global monitoring. Although numerous methods exist to extract phenological timing, in particular start-of-spring (SOS), from time series of reflectance data, a comprehensive intercomparison and interpretation of SOS methods has not been conducted. Here, we assess 10 SOS methods for North America between 1982 and 2006. The techniques include consistent inputs from the 8km Global Inventory Modeling and Mapping Studies Advanced Very High Resolution Radiometer NDVIg dataset, independent data for snow cover, soil thaw, lake ice dynamics, spring streamflow timing, over 16000 individual measurements of ground-based phenology, and two temperature-driven models of spring phenology. Compared with an ensemble of the 10 SOS methods, we found that individual methods differed in average day-of-year estimates by ! 60 days and in standard deviation by ! 20 days. The ability of the satellite methods to retrieve SOS estimates was highest in northern latitudes and lowest in arid, tropical, and Mediterranean ecoregions. The ordinal rank of SOS methods varied geographically, as did the relationships between SOS estimates and the cryospheric/hydrologic metrics. Compared with ground observations, SOS estimates were more related to the first leaf and first flowers expanding phenological stages. We found no evidence for time trends in spring arrival from ground- or model-based data; using an ensemble estimate from two methods that were more closely related to ground observations than other methods, SOS

Intercomparison, interpretation, and assessment of spring phenology in North America estimated from remote sensing for 1982-2006 M I C H A E L A. W H I T E*, K I R S T E N M. DE BEURS w , K A M E L D I D A Nz, D AV I D W. I N O U Y E § ,

Accurate land cover information is required for science, monitoring, and reporting. Land cover changes naturally over time, as well as a result of anthropogenic activities. Monitoring and mapping of land cover and land cover change in a consistent and robust manner over large areas is made possible with Earth Observation (EO) data. Land cover products satisfying a range of science and policy information needs are currently produced periodically at different spatial and temporal scales. The increased availability of EO data—particularly from the Landsat archive (and soon to be augmented with Sentinel-2 data)—coupled with improved computing and storage capacity with novel image compositing approaches, have resulted in the availability of annual, large-area, gap-free, surface reflectance data products. In turn, these data products support the development of annual land cover products that can be both informed and constrained by change detection outputs. The inclusion of time series change in the land cover mapping process provides information on class stability and informs on logical class transitions (both temporally and categorically). In this review, we present the issues and opportunities associated with generating and validating time-series informed annual, large-area, land cover products, and identify methods suited to incorporating time series information and other novel inputs for land cover characterization.

Optical remotely sensed time series data for land cover classification: A review

Recent advances in airborne and spaceborne hyperspectral imaging technology have provided end users with rich spectral, spatial, and temporal information. They have made a plethora of applications feasible for the analysis of large areas of the Earth?s surface. However, a significant number of factors-such as the high dimensions and size of the hyperspectral data, the lack of training samples, mixed pixels, light-scattering mechanisms in the acquisition process, and different atmospheric and geometric distortions-make such data inherently nonlinear and complex, which poses major challenges for existing methodologies to effectively process and analyze the data sets. Hence, rigorous and innovative methodologies are required for hyperspectral image (HSI) and signal processing and have become a center of attention for researchers worldwide.

/pdf/advances-in-hyperspectral-image-and-signal-processing-a-4kxt4zjyeh.pdf

Advances in Hyperspectral Image and Signal Processing: A Comprehensive Overview of the State of the Art

Advances in computing technology have fostered the development of new and powerful deep learning (DL) techniques, which have demonstrated promising results in a wide range of applications. Particularly, DL methods have been successfully used to classify remotely sensed data collected by Earth Observation (EO) instruments. Hyperspectral imaging (HSI) is a hot topic in remote sensing data analysis due to the vast amount of information comprised by this kind of images, which allows for a better characterization and exploitation of the Earth surface by combining rich spectral and spatial information. However, HSI poses major challenges for supervised classification methods due to the high dimensionality of the data and the limited availability of training samples. These issues, together with the high intraclass variability (and interclass similarity) –often present in HSI data– may hamper the effectiveness of classifiers. In order to solve these limitations, several DL-based architectures have been recently developed, exhibiting great potential in HSI data interpretation. This paper provides a comprehensive review of the current-state-of-the-art in DL for HSI classification, analyzing the strengths and weaknesses of the most widely used classifiers in the literature. For each discussed method, we provide quantitative results using several well-known and widely used HSI scenes, thus providing an exhaustive comparison of the discussed techniques. The paper concludes with some remarks and hints about future challenges in the application of DL techniques to HSI classification. The source codes of the methods discussed in this paper are available from: https://github.com/mhaut/hyperspectral_deeplearning_review .

Deep learning classifiers for hyperspectral imaging: A review

Based on annual average PM2.5 gridded dataset, this study first analyzed the spatiotemporal pattern of PM2.5 across Mainland China during 1998–2012. Then facilitated with meteorological site data, land cover data, population and Gross Domestic Product (GDP) data, etc., the contributions of latent geographic factors, including socioeconomic factors (e.g., road, agriculture, population, industry) and natural geographical factors (e.g., topography, climate, vegetation) to PM2.5 were explored through Geographically Weighted Regression (GWR) model. The results revealed that PM2.5 concentrations increased while the spatial pattern remained stable, and the proportion of areas with PM2.5 concentrations greater than 35 μg/m3 significantly increased from 23.08% to 29.89%. Moreover, road, agriculture, population and vegetation showed the most significant impacts on PM2.5. Additionally, the Moran’s I for the residuals of GWR was 0.025 (not significant at a 0.01 level), indicating that the GWR model was properly specified. The local coefficient estimates of GDP in some cities were negative, suggesting the existence of the inverted-U shaped Environmental Kuznets Curve (EKC) for PM2.5 in Mainland China. The effects of each latent factor on PM2.5 in various regions were different. Therefore, regional measures and strategies for controlling PM2.5 should be formulated in terms of the local impacts of specific factors.

/pdf/spatiotemporal-pattern-of-pm-2-5-concentrations-in-mainland-4wfpg0p6ne.pdf

Spatiotemporal Pattern of PM 2.5 Concentrations in Mainland China and Analysis of Its Influencing Factors using Geographically Weighted Regression

Most machine learning tasks can be categorized into classification or regression problems. Regression and classification models are normally used to extract useful geographic information from observed or measured spatial data, such as land cover classification, spatial interpolation, and quantitative parameter retrieval. This paper reviews the progress of four advanced machine learning methods for spatial data handling, namely, support vector machine (SVM)-based kernel learning, semi-supervised and active learning, ensemble learning, and deep learning. These four machine learning modes are representative because they improve learning performances from different views, for example, feature space transform and decision function (SVM), optimized uses of samples (semi-supervised and active learning), and enhanced learning models and capabilities (ensemble learning and deep learning). For spatial data handling via machine learning that can be improved by the four machine learning models, three key elements are learning algorithms, training samples, and input features. To apply machine learning methods to spatial data handling successfully, a four-level strategy is suggested: experimenting and evaluating the applicability, extending the algorithms by embedding spatial properties, optimizing the parameters for better performance, and enhancing the algorithm by multiple means. Firstly, the advances of SVM are reviewed to demonstrate the merits of novel machine learning methods for spatial data, running the line from direct use and comparison with traditional classifiers, and then targeted improvements to address multiple class problems, to optimize parameters of SVM, and to use spatial and spectral features. To overcome the limits of small-size training samples, semi-supervised learning and active learning methods are then utilized to deal with insufficient labeled samples, showing the potential of learning from small-size training samples. Furthermore, considering the poor generalization capacity and instability of machine learning algorithms, ensemble learning is introduced to integrate the advantages of multiple learners and to enhance the generalization capacity. The typical research lines, including the combination of multiple classifiers, advanced ensemble classifiers, and spatial interpolation, are presented. Finally, deep learning, one of the most popular branches of machine learning, is reviewed with specific examples for scene classification and urban structural type recognition from high-resolution remote sensing images. By this review, it can be concluded that machine learning methods are very effective for spatial data handling and have wide application potential in the big data era.

Advances of Four Machine Learning Methods for Spatial Data Handling: a Review

Recently, multifeature learning in collaborative representation classification (CRC) for hyperspectral images has generated promising performance. In this paper, two novel multifeature learning algorithms that update dictionary directly and indirectly are proposed. In order to offer the complementarity of multifeature, four different types of features—global feature (i.e., Gabor feature), local feature (i.e., local binary pattern), shape feature (i.e., extended multiattribute profiles), and spectral feature—are adopted in this paper. Under the hypothesis that most of the features should share the same coding pattern in CRC, this paper proposes to learn proper dictionaries for each feature until obtaining stable codes in a linear classifier. Furthermore, to avoid the explicit mapping of infinite-dimensional dictionaries in a nonlinear kernelized classifier, an indirect approach to construct the transformation matrix from original dictionaries to learn new dictionaries is developed. Three real hyperspectral images acquired from different sensors are adopted for performance evaluation. The experimental results demonstrate that the proposed methods can provide superior performance compared with those of the state-of-the-art classifiers.

Multifeature Dictionary Learning for Collaborative Representation Classification of Hyperspectral Imagery

The objective of this study is to classify the land cover types and analyze the land cover trend by incorporating phenological variability throughout a range of natural ecosystems using time-series remotely sensed images. First, a breaks for additive seasonal and trend (BFAST) approach is used to extract the phenology information from the time series. Second, a dynamic time warping (DTW) approach is adopted to screen the additional interpreted samples used for training. Third, some ensemble learning classifiers and the support vector machine (SVM) are performed to classify the land cover types based on the BFAST-derived phenology components. Finally, some inter-annual phenological markers are extracted to facilitate the land cover trend analysis by taking the climate fluctuations and anthropogenic forcing into consideration. The experimental results with normalized difference vegetation index (NDVI) and enhanced vegetation index (EVI) time-series data collected by the Moderate Resolution Imaging Spectrometer (MODIS) indicate that the classification accuracy is significantly improved by using the phenology information and the phenological markers can lead to a better understanding of the regional land cover change.

Phenology-Driven Land Cover Classification and Trend Analysis Based on Long-term Remote Sensing Image Series

Recently, deep learning (DL) is of great interest in hyperspectral image (HSI) classification. Although many effective frameworks exist in the literature, the generally limited availability of training samples poses great challenges in applying DL to HSI classification. In this paper, we present a novel DL framework, namely, semisupervised stacked autoencoders (Semi-SAEs) with cotraining, for HSI classification. First, two SAEs are pretrained based on the hyperspectral features and the spatial features, respectively. Second, fine-tuning is alternatively conducted for the two SAEs in a semisupervised cotraining fashion, where the initial training set is enlarged by designing an effective region growing method. Finally, the classification probabilities obtained by the two SAEs are fused using a Markov random field model solved by iterated conditional modes. Experimental results based on three popular hyperspectral data sets demonstrate that the proposed method outperforms other state-of-the-art DL methods.

Zhaohui Xue

Papers

Spatiotemporal Pattern of PM 2.5 Concentrations in Mainland China and Analysis of Its Influencing Factors using Geographically Weighted Regression

Advances of Four Machine Learning Methods for Spatial Data Handling: a Review

Multifeature Dictionary Learning for Collaborative Representation Classification of Hyperspectral Imagery

Phenology-Driven Land Cover Classification and Trend Analysis Based on Long-term Remote Sensing Image Series

Semisupervised Stacked Autoencoder With Cotraining for Hyperspectral Image Classification