Statistics for Spatial Data.

1. Introduction and Overview. 2. Detecting Influential Observations and Outliers. 3. Detecting and Assessing Collinearity. 4. Applications and Remedies. 5. Research Issues and Directions for Extensions. Bibliography. Author Index. Subject Index.

Regression Diagnostics: Identifying Influential Data and Sources of Collinearity

Climate change 2014 - Synthesis Report

We evaluate 179 classifiers arising from 17 families (discriminant analysis, Bayesian, neural networks, support vector machines, decision trees, rule-based classifiers, boosting, bagging, stacking, random forests and other ensembles, generalized linear models, nearest-neighbors, partial least squares and principal component regression, logistic and multinomial regression, multiple adaptive regression splines and other methods), implemented in Weka, R (with and without the caret package), C and Matlab, including all the relevant classifiers available today. We use 121 data sets, which represent the whole UCI data base (excluding the large-scale problems) and other own real problems, in order to achieve significant conclusions about the classifier behavior, not dependent on the data set collection. The classifiers most likely to be the bests are the random forest (RF) versions, the best of which (implemented in R and accessed via caret) achieves 94.1% of the maximum accuracy overcoming 90% in the 84.3% of the data sets. However, the difference is not statistically significant with the second best, the SVM with Gaussian kernel implemented in C using LibSVM, which achieves 92.3% of the maximum accuracy. A few models are clearly better than the remaining ones: random forest, SVM with Gaussian and polynomial kernels, extreme learning machine with Gaussian kernel, C5.0 and avNNet (a committee of multi-layer perceptrons implemented in R with the caret package). The random forest is clearly the best family of classifiers (3 out of 5 bests classifiers are RF), followed by SVM (4 classifiers in the top-10), neural networks and boosting ensembles (5 and 3 members in the top-20, respectively).

/pdf/do-we-need-hundreds-of-classifiers-to-solve-real-world-1igtahh9y9.pdf

Do we need hundreds of classifiers to solve real world classification problems

A wide range of methods for analysis of airborne- and satellite-derived imagery continues to be proposed and assessed. In this paper, we review remote sensing implementations of support vector machines (SVMs), a promising machine learning methodology. This review is timely due to the exponentially increasing number of works published in recent years. SVMs are particularly appealing in the remote sensing field due to their ability to generalize well even with limited training samples, a common limitation for remote sensing applications. However, they also suffer from parameter assignment issues that can significantly affect obtained results. A summary of empirical results is provided for various applications of over one hundred published works (as of April, 2010). It is our hope that this survey will provide guidelines for future applications of SVMs and possible areas of algorithm enhancement.

https://www.aboutgis.com/Publications/Mountrakis_SVM_review_in_remote_sensing_ISPRS2010.pdf

Support vector machines in remote sensing: A review

Comparing helicopter-borne profiling radar with airborne laser scanner data for forest structure estimation.

Trajectory estimation refers to the task of obtaining position and orientation estimates by fusing various sensor inputs. In kinematic mapping, global navigation satellite systems (GNSS) and inertial navigation systems (INS) are traditionally used to compute a trajectory which then serves as basis for direct or integrated orientation of the imaging sensors. As an inherently interdisciplinary problem, literature on trajectory estimation is broad. Apart from remote sensing itself, many recent advances come from autonomous navigation and robotics. This paper aims to provide a unified view of trajectory estimation with a focus on its role in kinematic mapping, specifically on the integration of GNSS, INS, laser scanners and cameras, as well as a survey of the related literature. Recent trends and challenges in trajectory estimation are identified and discussed. 

Integrated trajectory estimation for 3D kinematic mapping with GNSS, INS and imaging sensors: A framework and review

Zusammenfassung Der vorliegende Beitrag beschreibt die Konzeption und die methodischen Schritte des Projekts HORA 3.0, in dem Hochwasserrisikoflächen für ganz Österreich berechnet wurden. Die Analyseschritte umfassen: Qualitätskontrolle und Korrektur des Gewässernetzes und der Einzugsgebietsgrenzen; Berechnung der Abflussscheitel und Frachten gegebener Jährlichkeit; Erstellung eines digitalen Höhenmodells, das mit allen relevanten Hochwasserinformationen, einschließlich der Flussbettgeometrie, übereinstimmt; instationäre, zweidimensionale Simulation der Überflutungsflächen mit konsistenter Jährlichkeit. In jedem Schritt werden automatische und manuelle Verfahren kombiniert, um die lokalen hydrologischen und hydraulischen Bedingungen in einem bundesweiten Projekt möglichst zutreffend abzubilden. Die Hochwasserrisikoflächen mit einer Auflösung von 2 m für eine Gesamtflusslänge von 32.000 km sind bereits auf der HORA-Plattform ( www.hora.gv.at ) veröffentlicht. Das entwickelte Instrumentarium kann für weiterführende Projekte verwendet werden, wie z. B. für Visualisierungen, Schadensauswertungen und in Zukunft für die Berechnung pluvialer Hochwassergefahren. 

/pdf/hochwasserrisikozonierung-austria-3-0-hora-3-0-2cbjxnpc.pdf

HOchwasserRisikozonierung Austria 3.0 (HORA 3.0)

. Alpine rivers have experienced considerable changes in channel morphology over the last century. Natural factors and human disturbance are the main drivers of changes in channel morphology that modify natural sediment and flow regimes at local, catchment, and regional scales. In glaciated catchments, river sediment loads are likely to increase due to increasing 15 snow and glacier melt runoff, facilitated by climate changes. Additionally, channel erosion and depositional dynamics and patterns are influenced by sediment delivery from hillslopes, and sediment in the forefields of retreating glaciers. In order to reliably assess the magnitudes of the channel-changing processes and their frequencies due to recent climate change, the investigation period needs to be extended to the last century, ideally back to the end of the Little Ice Age. Moreover, a high temporal resolution is required to account for the history of changes in channel morphology and for better detection and 20 interpretation of related processes. The increasing availability of digitized historical aerial images and advancements in digital photogrammetry provides the basis for reconstructing and assessing the long-term evolution of the surface, both in terms of planimetric mapping and the generation of historical digital elevation models (DEMs). The

Evolution of an Alpine proglacial river during seven decades of deglaciation quantified from photogrammetric and LiDAR digital elevation models

In this paper we investigate the determination of camera relative orientation in videos from time of flight (ToF) range imaging camera. The task of estimating the relative orientation is realized by fusion of range flow and optical flow constraints, which integrates the range and the intensity channels in a single framework. We demonstrate our approach on videos from a ToF camera involving camera translation and rotational motion and compare it with the ground truth data. Furthermore we distinguish camera motion from an independently moving object using a robust adjustment.

Norbert Pfeifer

Papers

Comparing helicopter-borne profiling radar with airborne laser scanner data for forest structure estimation.

Integrated trajectory estimation for 3D kinematic mapping with GNSS, INS and imaging sensors: A framework and review

HOchwasserRisikozonierung Austria 3.0 (HORA 3.0)

Evolution of an Alpine proglacial river during seven decades of deglaciation quantified from photogrammetric and LiDAR digital elevation models

Relative orientation of videos from range imaging cameras