Refine
Document Type
- Bachelor Thesis (3) (remove)
Has Fulltext
- yes (3)
Is part of the Bibliography
- no (3)
Keywords
- Deep learning (3) (remove)
Institute
Training and evaluating deep learning models on road graphs for traffic prediction using SUMO
(2024)
The escalation of traffic volume in urban areas poses multifaceted challenges including increased accident risks, congestion, and prolonged travel times. Traditional approaches of expanding road infrastructure face limitations such as space constraints and the potential exacerbation of traffic issues.
Intelligent Transport Systems (ITS) present an alternative strategy to alleviate traffic problems by leveraging data-driven solutions. Central to ITS is traffic prediction, a process vital for applications like Traffic Management and Navigation Systems.
Recent advancements in traffic prediction have witnessed a surge of interest, particularly in deep learning methods optimized for graph-based data processing, being considered the most promising avenue presently.
These methods typically rely on real-life datasets containing traffic sensor data such as METR-LA and PeMS. However, the finite nature of real-life data prompts exploration into augmenting training and testing datasets with simulated traffic data.
This thesis explores the potential of utilizing traffic simulations, employing the microscopic traffic simulator SUMO, to train and test deep learning models for traffic prediction. A framework integrating PyTorch and SUMO is proposed for this purpose, aiming to elucidate the feasibility and effectiveness of using simulated traffic data for enhancing predictive models in traffic management systems.
Recent developments in the field of deep learning have shown promising advances for a wide range of historically difficult computer vision problems. Using advanced deep learning techniques, researchers manage to perform high-quality single-image super-resolution, i.e., increasing the resolution of a given image without major losses in image quality, usually encountered when using traditional approaches such as standard interpolation. This thesis examines the process of deep learning super-resolution using convolutional neural networks and investigates whether the same deep learning models can be used to increase OCR results for low-quality text images.
Bis heute ist völlig unbekannt, ob wir allein im Universum sind. Um auf dieses Thema eine Antwort zu finden, überprüft diese Bachelorarbeit, ob Convolutional (CNN) und Recurrent Neural Networks (RNN) für die Erkennung außerirdischer Signale geeignet sind.
Das Ziel war dabei, in einem Datensatz bestehend aus Spektrogrammen mehr als 50% aller außerirdischer Signale zu erkennen, da nur so ein Neuronales Netzwerk ein besseres Resultat als eine zufällige Klassifikation liefert, bei der im Mittel 50% aller Signale erkannt werden.
Dabei zeigte sich, dass sich mit beiden Varianten der Neuronalen Netzwerke bis zu 90% aller Signale erkennen lassen, die Vorhersagen von CNNs allerdings verlässlicher sind. RNNs bieten hingegen aufgrund ihrer geringeren Größe einen deutlich leichtgewichtigeren Ansatz und führen zu einer signifikanten Speicherersparnis.
Daraus folgt, dass Neuronale Netzwerke bei der Suche nach außerirdischem Leben im Universum helfen können, um die Frage „Sind wir allein im Universum?“ endgültig zu beantworten.