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

In this paper, we address various challenges in multi-pedestrian and vehicle tracking in high-resolution aerial imagery by intensive evaluation of a number of traditional and Deep Learning based Single- and Multi-Object Tracking methods. We also describe our proposed Deep Learning based Multi-Object Tracking method Aerial MPTNet that fuses appearance, temporal, and graphical information using a Siamese Neural Network, a Long Short-Term Memory, and a Graph Convolutional Neural Network module for more accurate and stable tracking. Moreover, we investigate the influence of the Squeeze-and-Excitation layers and Online Hard Example Mining on the performance of Aerial MPTNet. To the best of our knowledge, we are the first to use these two for regression-based Multi-Object Tracking. Additionally, we studied and compared the L1 and Huber loss functions. In our experiments, we extensively evaluate AerialMPTNet on three aerial Multi-Object Trackingdatasets, namely AerialMPT and KIT AIS pedestrian and vehicle datasets. Qualitative and quantitative results show that AerialMPTNet outperforms all previous methods for the pedestrian datasets and achieves competitive results for the vehicle dataset. In addition, Long Short-Term Memory and Graph Convolutional Neural Network modules enhance the tracking performance. Moreover, using Squeeze-and-Excitation and Online Hard Example Mining significantly helps for some cases while degrades the results for other cases. In addition, according to the results, L1 yields better results with respect to Huber loss for most of the scenarios. The presented results provide a deep insight into challenges and opportunities of the aerial Multi-Object Tracking domain, paving the way for future research.

Multiple Pedestrians and Vehicles Tracking in Aerial Imagery Using a Convolutional Neural Network

Accurate classification of hyperspectral data is still a competitive task and new classification methods are developed to achieve desired tasks of hyperspectral data use. The objective of this paper is to develop a new method for hyperspectral data classification ensuring the classification model properties like transferability, generalization, probabilistic interpretation, etc. While factor graphs (undirected graphical models) are unfortunately not widely employed in remote sensing tasks, these models possess important properties such as representation of complex systems to model estimation/decision making tasks. In this paper we present a new method for hyperspectral data classification using factor graphs. Factor graph (a bipartite graph consisting of variables and factor vertices) allows factorization of a more complex function leading to definition of variables (employed to store input data), latent variables (allow to bridge abstract class to data), and factors (defining prior probabilities for spectral features and abstract classes; input data mapping to spectral features mixture and further bridging of the mixture to an abstract class). Latent variables play an important role by defining two-level mapping of the input spectral features to a class. Configuration (learning) on training data of the model allows calculating a parameter set for the model to bridge the input data to a class. The classification algorithm is as follows. Spectral bands are separately pre-processed (unsupervised clustering is used) to be defined on a finite domain (alphabet) leading to a representation of the data on multinomial distribution. The represented hyperspectral data is used as input evidence (evidence vector is selected pixelwise) in a configured factor graph and an inference is run resulting in the posterior probability. Variational inference (Mean field) allows to obtain plausible results with a low calculation time. Calculating the posterior probability for each class and comparison of the probabilities leads to classification. Since the factor graphs operate on input data represented on an alphabet (the represented data transferred into multinomial distribution) the number of training samples can be relatively low. Classification assessment on Salinas hyperspectral data benchmark allowed to obtain a competitive accuracy of classification. Employment of training data consisting of 20 randomly selected points for a class allowed to obtain the overall classification accuracy equal to 85.32% and Kappa equal to 0.8358. Representation of input data on a finite domain discards the curse of dimensionality problem allowing to use large hyperspectral data with a moderately high number of bands.

/pdf/hyperspectral-data-classification-using-factor-graphs-53lejrzo9g.pdf

Hyperspectral data classification using factor graphs

. Geo-referenced real-time vehicle and person tracking in aerial imagery has a variety of applications such as traffic and large-scale event monitoring, disaster management, and also for input into predictive traffic and crowd models. However, object tracking in aerial imagery is still an unsolved challenging problem due to the tiny size of the objects as well as different scales and the limited temporal resolution of geo-referenced datasets. In this work, we propose a new approach based on Convolutional Neural Networks (CNNs) to track multiple vehicles and people in aerial image sequences. As the large number of objects in aerial images can exponentially increase the processing demands in multiple object tracking scenarios, the proposed approach utilizes the stack of micro CNNs, where each micro CNN is responsible for a single-object tracking task. We call our approach Stack of Micro-Single- Object-Tracking CNNs (SMSOT-CNN). More precisely, using a two-stream CNN, we extract a set of features from two consecutive frames for each object, with the given location of the object in the previous frame. Then, we assign each MSOT-CNN the extracted features of each object to predict the object location in the current frame. We train and validate the proposed approach on the vehicle and person sets of the KIT AIS dataset of object tracking in aerial image sequences. Results indicate the accurate and time-efficient tracking of multiple vehicles and people by the proposed approach.

/pdf/multiple-vehicles-and-people-tracking-in-aerial-imagery-4azgyx1wgd.pdf

Multiple vehicles and people tracking in aerial imagery using stack of micro single-object-tracking cnns

MOMS (Modular Optoelectronic Multispectral Stereo Scanner) is a German spaceborne pushbroom scanner for high resolution (HR), multispectral (MS) and threefold along-track stereoscopic imaging. The camera was first flown on the second German Space Shuttle Mission in spring 1993. After the 11 days shuttle mission the camera system was refurbished for a second flight onboard the PRIRODA module of the Russian space station MIR. This mission, called MOMS-2P, lasted from May 1996 to August 1999. MOMS-2P was equipped with a navigation package consisting of a GPS receiver and two redundant gyro subsystems for obtaining position and attitude data with high accuracy. For the MIR orbit altitude of 400 km the ground pixel size was 18 m and 6 m (HR). The orbit inclination was 51.6°. The data takes mainly cover Europe, North-East Africa, Middle East and South America, but also selected image strips from other parts in the world are available. In total approximately 65 Mio. km2 of the land mass have been recorded. In the meantime a series of MOMS data takes have been evaluated. Accuracy analysis using independent ground control led to point accuracies of better than 10 m in X, Y, Z for test sites in Germany. The geometric calibration was conducted by band to band registration for the multispectral channels and by photogrammetric adjustment using precise ground control for the stereo channels (inverse photogrammetry). The paper decribes the imaging technique of the camera, presents the actual status of data recording and data availability, reports on the applied calibration methods and summarises results of photogrammetric data evaluation.

MOMS - The First Along Track Stereo Camera in Space - A Mission Review

For the radiometric calibration of the MOMS-2P camera on board of the Russian MIR-station an inflight sun calibration method using a diffuser plate in front of the optics is applied. By tilting the diffuser plate by 20 degrees with respect to the entrance optics the direct sunlight is reflected into the optics. Prior to the absolute radiometric calibration of every pixel the image has to be corrected for relative radiometric disturbances along the scan lines. The reasons are the odd/even effect and the CCD typical non-uniformity. To correct for these effects relative calibration tables, derived from the diffuser plate calibration, are applied to the image. The applicability and stability of these tables is only given to an accuracy of about 1-2% in the nominal case. Therfore the remaining disturbances, probably caused by deficiencies of the internal electronic amplifier circuits and temperature variations, are removed using empirical methods. To correct for these random type of image failures, a digital filter in the spatial domain was developed. It is based on an averaging technique that acts in a large kernel size. Since there seems not to be complete linearity for very high and very low brightness levels, these levels have to be excluded from the averaging procedure. The image contrast and information content is less influenced than with other known filter techniques. In this paper the procedure to derive the level L1A Data in shown. After correcting for the relative radiometric differences the measured solar irradiance values can also be used for absolute calibration because the extraterrestrial spectrum of the sun is very well known. The methods to derive the absolute calibration factors are described and first results are presented.

/pdf/radiometric-calibration-of-the-moms-2p-camera-4zvw18rekt.pdf

Peter Reinartz

Papers

Multiple Pedestrians and Vehicles Tracking in Aerial Imagery Using a Convolutional Neural Network

Hyperspectral data classification using factor graphs

Multiple vehicles and people tracking in aerial imagery using stack of micro single-object-tracking cnns

MOMS - The First Along Track Stereo Camera in Space - A Mission Review

Radiometric calibration of the MOMS-2P camera