• Zielgruppen
  • Suche
 

Big Data and Machine Learning

Object detection in airborne laser scanning (ALS) data using deep learning

Bild zum Projekt Object detection in airborne laser scanning (ALS) data using deep learning

Researcher:

Kazimi, Thiemann, Sester

Funded by:

MWK Pro*Niedersachsen

Brief description:

In partnership with the Lower Saxony State Office for Preservation of Historic Monuments, we are developing a method for automatically detecting archaeological objects in airborne laser scanning data. The type of objects to be detected are mainly those of interest by archaeologists such as heaps, shafts, charcoal piles, pits, barrows, bomb craters, hollow ways, etc. They could be point, linear, or areal objects. To this end, we are using deep learning techniques; namely, convolutional neural networks (CNNs) to classify height images from the region of interest. A combination of multiple (in most cases 5) CNN classifiers are then used to detect and localize objects of interest in a digital terrain model acquired from the region of interest.

 

| details |

 

Ja, wo laufen sie denn?

Bild zum Projekt Ja, wo laufen sie denn?

Researcher:

Feuerhake, Sester

Brief description:

Für Profi-Trainer oder auch einfache Hobby-Kicker. Vielen Fußballbegeisterten wird der Weg zum Taktikfuchs durch eine automatisierte Spielanalyse am Computer erleichtert. Ausgeklügelte Verfahren ermöglichen eine einfachere Bewertung der Leistung der Akteure.

 

| details |

 

Real Time Prediction of Pluvial Floods and Induced Water Contamination in Urban Areas (EVUS)

Bild zum Projekt Real Time Prediction of Pluvial Floods and Induced Water Contamination in Urban Areas (EVUS)

Researcher:

Feng, Sester

Funded by:

BMBF Georisiken

Brief description:

This project aims at developing a fast forecast model for pluvial floods in the city of Hannover. The main goal of the subproject for ikg is to integrate new sensors for the flood prediction models.

 

| details |

 

Scene analysis – Pattern recognition on pedestrian trajectories

Bild zum Projekt Szenenanalyse - Mustererkennung in Personentracks

Researcher:

Kuntzsch, Sester

Brief description:

The project deals with an automated detection of salient movement patterns within pedestrian trajectories, given as sequence of time-discrete observations of object positions. Patterns, that imply unusual behavior, are defined beforehand and techniques for their automated detection are developed. This includes qualitative results with regards to detection rates. Patterns may be defined in different context: we distinguish between individual and group patterns evolving from multiple individual movements within a common spatio-temporal context. Through the entire analysis process it is mandatory to take influences of the context itself on the movements into account, e.g. a trajectory may be influenced by stationary or mobile obstacles within the traversed space or typical movement patterns may vary at different times of the day. For detection of security-critical, human-induced behavior within a single scenario typical movement patterns within spatio-temporal context are learned from observations. This allows anomaly detection mechanisms to identify unusual movement behavior, which may imply security critical behavior, in near real-time.

 

| details |

 

3D-Objektextraktion aus hochaufgelösten 3d-Punktwolken

Bild zum Projekt 3D-Objektextraktion aus hochaufgelösten 3d-Punktwolken

Researcher:

Politz, Sester

Funded by:

Landesvermessungsämter Niedersachsen, Schleswig-Holstein, Mecklenburg-Vorpommern

Brief description:

In den Landesvermessungsbehörden liegen flächendeckende, kontrollierte Airborne Lascerscanning-Datensätze mit unterschiedlichen Punktdichten vor, welche i.d.R. mindestens in die Klassen Boden- und Nichtbodenpunkte differenziert wurden. In der Arbeitsgemeinschaft der Vermessungsverwaltungen (AdV) wird ein Aktualisierungszyklus von 10 Jahren diskutiert. Weiterhin leiten die Landesvermessungsämter auf Basis von digitalen Bildflügen mit hohen Überlappungen 3D-Punktwolken mit dem sogenannten „Dense-Image-Matching“-Verfahren (DIM) ab, welche eine Auflösung im Pixelbereich besitzen. Radiometrische Information aus den Luftbildern ergänzen die Informationstiefe dieser Punktwolken, welche aufgrund der Bildkorrelation in der Regeln auf ein Oberflächenmodell begrenzt sind. Hierbei ist ein 2-3-jähriger Befliegungszyklus die Basis.

 

| details |

 

Big Data and Machine Learning completed

RainCars

Bild zum Projekt RainCars

Researcher:

Fitzner, Sester

Brief description:

This idea would be easily technically feasible if the cars are provided with GPS and a small memory chip for recording the coordinates, car speed and wiper frequency. This initial research will explore theoretically the benefits of such an approach. For that valid relationships between wiper speed and rainfall rate (W-R relationship) are assumed and derived from laboratory and field experiments. Different traffic models are developed to generate motorcars on roads in a river basin. Radar data are used as reference truth rainfall fields. Rainfall from these fields is sampled from the conventional rain gauge and dynamic car networks. Areal rainfall is calculated from these networks for different scales using geostatistical interpolation methods and compared against truth radar data. The car sensors can be considered as a geosensor network. It allows to measure and process information locally in a decentralized way and thus has benefits with respect to scalability, which is crucial when large areas have to be covered with large amounts of measurement units.

 

| details |

 

Anomalous Pattern Detection from GPS-Trajectories

Bild zum Projekt Anomalous Pattern Detection from GPS-Trajectories

Brief description:

The anomalous pattern detection is of great interest for the applications in the areas of navigation/driver assistant system, surveillance and emergency management. In this work we focus on the GPS-Trajectories finding where the driver is encountering navigation problems.

 

| details |

 

Q-Trajectories - Dezentrale Bestimmung von Bewegungsmustern aus Trajektorien

Bild zum Projekt Q-Trajectories - Dezentrale Bestimmung von Bewegungsmustern aus Trajektorien

Brief description:

Ziel dieses Teilprojekts ist die Erkennung und Bewertung von Bewegungsmustern in Trajektorien mit Hilfe effizienter, dezentraler Analysemethoden. Dabei sollen Auffälligkeiten und kritische Verhaltensweisen ausfindig gemacht werden. Anwendungsmöglichkeiten für ein derartiges Verfahren könnten u.a. größere Sensor-/Kameranetze zur Überwachung von Menschenmengen (z.B. Stau, Gruppenverhalten, …) oder zur Beobachtung von Verhalten von Tieren sein.

 

| details |

 

Mining GPS-Trajectory Data for Map Refinement and Behavior Detection

Bild zum Projekt Mining GPS-Trajectory Data for Map Refinement and Behavior Detection

Brief description:

In today’s world, we have increasingly sophisticated means to record the movement of moving objects such as vehicles, humans and animals in the form of spatio-temporal trajectory data. As a consequence of this development, increasing volumes of such data are being accumulated at an extremely fast rate. A trajectory is usually represented by an array of structured positions in space and time, i.e. each has a signature of specific location (geospatial coordinate tags) in time (time stamp tags).

 

| details |