There is, I think, something ethereal about i —the square root of minus one. I remember first hearing about it at school. It seemed an odd beast at that time—an intruder hovering on the edge of reality.

Usually familiarity dulls this sense of the bizarre, but in the case of i it was the reverse: over the years the sense of its surreal nature intensified. It seemed that it was impossible to write mathematics that described the real world in …

I and i

Machine Learning is the study of methods for programming computers to learn. Computers are applied to a wide range of tasks, and for most of these it is relatively easy for programmers to design and implement the necessary software. However, there are many tasks for which this is difficult or impossible. These can be divided into four general categories. First, there are problems for which there exist no human experts. For example, in modern automated manufacturing facilities, there is a need to predict machine failures before they occur by analyzing sensor readings. Because the machines are new, there are no human experts who can be interviewed by a programmer to provide the knowledge necessary to build a computer system. A machine learning system can study recorded data and subsequent machine failures and learn prediction rules. Second, there are problems where human experts exist, but where they are unable to explain their expertise. This is the case in many perceptual tasks, such as speech recognition, hand-writing recognition, and natural language understanding. Virtually all humans exhibit expert-level abilities on these tasks, but none of them can describe the detailed steps that they follow as they perform them. Fortunately, humans can provide machines with examples of the inputs and correct outputs for these tasks, so machine learning algorithms can learn to map the inputs to the outputs. Third, there are problems where phenomena are changing rapidly. In finance, for example, people would like to predict the future behavior of the stock market, of consumer purchases, or of exchange rates. These behaviors change frequently, so that even if a programmer could construct a good predictive computer program, it would need to be rewritten frequently. A learning program can relieve the programmer of this burden by constantly modifying and tuning a set of learned prediction rules. Fourth, there are applications that need to be customized for each computer user separately. Consider, for example, a program to filter unwanted electronic mail messages. Different users will need different filters. It is unreasonable to expect each user to program his or her own rules, and it is infeasible to provide every user with a software engineer to keep the rules up-to-date. A machine learning system can learn which mail messages the user rejects and maintain the filtering rules automatically. Machine learning addresses many of the same research questions as the fields of statistics, data mining, and psychology, but with differences of emphasis. Statistics focuses on understanding the phenomena that have generated the data, often with the goal of testing different hypotheses about those phenomena. Data mining seeks to find patterns in the data that are understandable by people. Psychological studies of human learning aspire to understand the mechanisms underlying the various learning behaviors exhibited by people (concept learning, skill acquisition, strategy change, etc.).

Machine learning

On robust estimation of the location parameter

A scheme is developed for classifying the types of motion perceived by a humanlike robot. It is assumed that the robot receives visual images of the scene using a perspective system model. Equations, theorems, concepts, clues, etc., relating the objects, their positions, and their motion to their images on the focal plane are presented. >

http://ieeexplore.ieee.org/iel2/815/3148/00100651.pdf

Robot vision

For 11 days in February 2000, the Shuttle Radar Topography Mission (SRTM) successfully recorded by interferometric synthetic aperture radar (InSAR) data of the entire land mass of the earth between 60°N and 57°S. The data acquired in C- and X-bands are processed into the first global digital elevation models (DEMs) at 1 arc sec resolution, by NASA-JPL and German aerospace center (DLR), respectively. From the perspective of the SRTM-X system, we give in this paper an overview of the mission and the DEM production, as well as an evaluation of the DEM product quality. Special emphasis is on challenges and peculiarities of the processing that arose from the unique design of the SRTM system, which has been the first single-pass interferometer in space.

The shuttle radar topography mission—a new class of digital elevation models acquired by spaceborne radar

One of the first essential steps in response to a crisis event is to provide current, precise, and rapid information about the extent of the event, the affected population and infrastructure (e.g., information for relief teams). Experiences in crisis mapping show, that remote sensing data are an important information source to get an overview of the situation. For an efficient use of geo-information before or during a crisis of any kind, all information should be orthorectified. 
Satellite data are delivered normally geocoded with varying accuracy. However, today orthorectification without ground control points (GCP) using only ephemeris and attitude data measured on board the satellite provide an absolute accuracy of about 20 m to 1 km (depending on the satellite). For the intended applications like change detection this accuracy is not sufficient. For this purpose, accuracy in the sub-pixel range is necessary requiring a precise orthorectification. Different methods deployed in different software packages are analyzed and compared in this paper. They are chosen depending on the available metadata.
To get an objective comparison of the different orthorectification methods, their accuracy is determined for several test datasets in this study. Additionally to the comparison of the accuracy, further criteria are analyzed like the potential of the algorithm for optimization and automation and the computing time needed, since the timely supply of information is also one of the most important requirements in crisis respond mapping.
Results using the commercially available software packages Erdas Imagine, PCI Geomatics and ENVI are compared to products of the in-house developed software XDibias of the German Aerospace Center (DLR). All programs result in high accuracy. Differences are experienced in the computing times and userfriendlyness. While the commercial programs are very user friendly, knowledge about the details of the in-house developed software are very helpful and the automation potential of it has already been proven.

Comparison of orthorectification methods suitable for rapid mapping using direct georeferencing and RPC for optical satellite data

Various deep learning applications benefit from multi-task learning with multiple regression and classification objectives by taking advantage of the similarities between individual tasks. This can result in improved learning efficiency and prediction accuracy for the task-specific models compared to separately trained models. In this paper, we make an observation of such influences for important remote sensing applications like elevation model generation and semantic segmentation tasks from the stereo half-meter resolution satellite digital surface models (DSMs). Mainly, we aim to generate good-quality DSMs with complete, as well as accurate level of detail (LoD)2-like building forms and to assign an object class label to each pixel in the DSMs. For the label assignment task, we select the roof type classification problem to distinguish between flat, non-flat, and background pixels. To realize those tasks, we train a conditional generative adversarial network (cGAN) with an objective
function based on least squares residuals and an auxiliary term based on normal vectors for further roof surface refinement. Besides, we investigate recently published deep learning architectures for both tasks and develop the final end-to-end network, which combines different models, as using
them first separately, they provide the best results for their individual tasks.

Multi-Task cGAN for Simultaneous Spaceborne DSM Refinement and Roof-Type Classification

For the generation of 3D city models from satellite stereo imagery beyond the generation of digital surface models (DSM) from stereo data the next crucial step is the separation of urban 3D objects from ground. To do this the most common method is the derivation of a so called digital terrain model (DTM) from the DSM. The DTM should ideally contain only the surface of the ground on which the urban objects are located. Since only the surface of the objects can be seen from space, sophisticated methods have to be developed to gain information of the bare ground. In this paper selected methods for the extraction of a DTM from a DSM are described and evaluated. The evaluation is done by applying the methods to synthetically generated DSMs. These synthetical DSMs are a combination of ground and typical urban objects put on top of it. The application of the DTM extraction methods should recover in turn the original ground model as good as possible. Also the sum of the obtained DTM and the profile of the urban objects should reconstruct the original DSM. The profile of the urban objects ist often referenced as normalized digital elevation model (nDEM). But in general the equation DSM = DTM + nDEM is not always valid ‐ especially for buildings situated on the slope of a hill. If the nDEM would simply be the difference of DSM ‐ DTM the slope of the hill ‐ contained in the DTM ‐ will be reflected on the roof of the buildings. So also an advanced method for derivation of the nDEM from DSM and DTM is presented and tested.

/pdf/evaluation-of-selected-methods-for-extracting-digital-x7nafvld4c.pdf

Evaluation of selected methods for extracting digital terrainmodels from satellite born digital surface models in urban areas

The rigorous approach for the derivation of digital surface models (DSM) and orthoimages from satellite and airborne scanner imagery via full modelling of the imaging process requires the knowledge of exterior and interior orientation of the camera. For IKONOS and QuickBird this basic data is not (fully) available to the user. Instead, rational polynomial functions (RPF) in standard form are provided as a substitute for describing the relationship between image and object space. To be able to derive DSM and orthoimages for this type of data even in the absence of appropriate ground control, software based on RPF has been developed. The software allows a lot of datums and projections for input and output data. This is important for worldwide application e.g. in case of catastrophes. The paper shows that a RPF is very near to linear in the object space variables longitude, latitude and height. This explains a large convergence radius and a rapid convergence in case of forward intersection. Residuals in image space from forward intersection are small and nearly constant in case of the GEO IKONOS-2 images tested. Large (also nearly constant) residuals in case of the QuickBird stereo pair can be used for a relative correction of the RPF either using residuals from verticality constraints with manually measured top/foot tie point pairs of large buildings or from forward intersection of a set of excellent tie points from matching. These corrected RPF can then be used to create a self-consistent set of DSM and orthoimages. The DSM derived from the high resolution data via RPF processing have been compared to best DSM available which was in most cases a C-Band SRTM model. Estimated mapping accuracy was found to correspond with the promises of the image provider in the case of available ground control.

DSM and Orthoimages from QuickBird and Ikonos Data using Rational Polynomial Functions

A fast detection and visualization of change in crisis areas is an important condition for planning and coordination of help. The availability of new satellites with high temporal resolution (e.g. RapidEye) and/or high spatial resolution (e.g. Quickbird) provides the basis for a better visualization of multitemporal change. For automated change detection, a large number of algorithms has been proposed and developed. This article describes the results of four texture based change detection approaches that were applied to satellite images of the Darfur crisis region. In our methodology we calculate firstly different texture characteristics (“energy”, “correlation”, “contrast” and “inverse distance moment”), for a whole image at two (or more) different times. The second step is to test the capability of known change detection methods (image-differencing, image-ratioing, regression analysis and principal component analysis) to visualize the change of settlement areas through these texture characteristics and texture images, respectively. The comparison of different texture characteristics with different change detection methods shows that best results can be obtained using a selective bitemporal principal component analysis with the texture feature “energy”.

/pdf/change-visualization-through-a-texture-based-analysis-3si62ox2hr.pdf

Peter Reinartz

Papers

Comparison of orthorectification methods suitable for rapid mapping using direct georeferencing and RPC for optical satellite data

Multi-Task cGAN for Simultaneous Spaceborne DSM Refinement and Roof-Type Classification

Evaluation of selected methods for extracting digital terrainmodels from satellite born digital surface models in urban areas

DSM and Orthoimages from QuickBird and Ikonos Data using Rational Polynomial Functions

Change Visualization through a Texture-Based Analysis Approach for Disaster Applications