Index

Automatic detection of Bronchoscopes on x-ray images

Scamming Scammers using Large Language Models

This Master Thesis is a cooperation with the Chair of Applied Cryptography.

Work description
In the digital age, scam emails have become a serious threat. These fraudulent emails aim to steal sensitive information or cause financial damage. This thesis aims to better understand the problem of scam emails and develop effective solutions to reduce their success. We will address several aspects, including the vulnerability of email addresses to scammers, the differentiation of scam emails from other dubious messages, the automation of responses through Large Language Models (LLMs), the detection of the usage of LLMs by the scammers, and the evaluation of the economic damage to the scammers based on the data obtained. We aim to strengthen the security of digital communication and help minimize the risks for users and organizations.

The following questions should be considered:

• How can an email address be made vulnerable to scammers?
• How can emails from scammers be distinguished from other dubious emails?
• How can LLM responses be automated and customized?
• How quickly do scammers recognize automated responses?
• How can we accurately assess the extent of the economic harm caused by the scammer using our collected data?

 

Prerequisites
Prerequisites for this task include good knowledge of Deep Learning and IT Security, familiarity with Python and PyTorch, and the capability to work independently.

For your application, please send your transcript of record.

Quantification of Metal Artifacts in Metal Artifact Avoidance

AI-based Pavement Recognition System for Vehicle Road Infrastructure

AI-Driven Monuments Identification System and its details

In the realm of computer vision, significant research is currently dedicated to object detection
and recognition. Research groups and developers are actively striving to enhance machine
learning solutions, aiming to boost the accuracy of image detection and recognition in
accordance with specific use cases. The German AI-driven Monuments Detection System is an
innovative project aimed at providing tourists with an enhanced experience by leveraging
artificial intelligence (AI) to recognize and provide historical information about prominent
monuments in Germany. This document presents the project’s methodology, results, and
implications. With this machine, a user can scan the historical place and view its historical
details. There are five categories used in this model. Python is used as a programming
language with the TensorFlow framework.

Synthetic data creation of defect images for CNN training using GAN

Master Thesis in Cooperation with Infineon Technologies AG

External Supervision:
Weichselbaumer Christoph (BE R UPE TEST) (Christoph.Weichselbaumer@infineon.com)

Working Tittle:
Synthetic data creation of defect images for CNN training using GAN

Research Question:
Can the use of Generative Adversarial Networks (GANs) for generating synthetic data of particles and scratches on a transparent background improve the accuracy of Convolutional Neural Networks (CNNs) in defect detection?

Description:
The proposed Master’s thesis aims to address the challenge of providing ground truth for model training in AI use-cases, specifically in the context of detecting defects in products. Currently, the process of manual review and labeling of images for CNN [1] training is highly time-consuming and costly, and the most critical and relevant defects are often the least present in real data due to the design of products to minimize such defects. Moreover, data shifts in production can also affect the training of models.
To overcome these challenges, the proposed Master’s thesis will focus on the use of Generative Adversarial Networks (GANs) [2],[3], [4] to generate synthetic data for the minority classes for CNN training. The goal of the thesis is to create one or more GANs that are capable of generating defect images of particles and scratches on a transparent background, and to evaluate the performance of these generated images by measuring their classification accuracy using an existing CNN [5].

The proposed research will involve two sub-targets:
1. Developing GANs capable of generating defect images of particles and scratches on a transparent background.
2. Creating GANs capable of image-to-image translation to generate defect images of particles and scratches.
The focus of the research will be on generating defect images on a transparent background, with image-to-image translation considered an optional target depending on the progress of the research.

The approach to using GANs involves using existing defect image datasets for training, and then generating synthetic images to fill gaps in the real data. The GANs will be trained to generate synthetic images of particles and scratches on a transparent background, and these images will be combined with a “golden die” background to create an enhanced dataset for CNN training.
The suggested topic has the potential to significantly reduce the time and cost associated with manual review and labeling of images for CNN training, and to provide a more diverse and relevant dataset for model training. The proposed approach can also be applied to other defect types, such as stains and chipping, to further enhance the dataset for CNN training.

Literature:
[1]. O’Shea, Keiron, and Ryan Nash. “An introduction to convolutional neural networks.” arXiv preprint arXiv:1511.08458 (2015).
[2]. Ali, Safinah, Daniella DiPaola, and Cynthia Breazeal. “What are GANs? introducing generative adversarial networks to middle school students.” Proceedings of the AAAI Conference on Artificial Intelligence. Vol. 35. No. 17. 2021.
[3]. Figueira, Alvaro, and Bruno Vaz. “Survey on synthetic data generation, evaluation methods and GANs.” Mathematics 10.15 (2022): 2733.
[4]. Eilertsen, Gabriel, et al. “Ensembles of GANs for synthetic training data generation.” arXiv preprint arXiv:2104.11797 (2021).
[5]. Buda, Mateusz, Atsuto Maki, and Maciej A. Mazurowski. “A systematic study of the class imbalance problem in convolutional neural networks.” Neural networks 106 (2018): 249-259.

End-to-end detection and 3D localization of implants from multi-view images for surgical CBCT metal artifact avoidance

This project investigates the feasibility of directly regressing pose and position of metallic implants from multiple calibrated X-Ray images. In the context of intraoperative Cone-beam Computed Tomography (CBCT) imaging, this information can be used to avoid metal artifacts by adapting the scanning trajectory such as discussed in this previous work [1].

By formulating the problem as a set prediction problem, we can build on previous works such as DETR [1] to design an algorithm which directly models the depicted metallic objects. In addition to applying these existing works, which were developed for and tested on day-to-day optical images, we assess the possibility of incorporating additional knowledge about the relative geometry between the X-Ray images into the model architecture.

[1] Cone-beam CT trajectory optimization for metal artifact avoidance using ellipsoidal object parameterizations (spiedigitallibrary.org)
[2] [2005.12872] End-to-End Object Detection with Transformers (arxiv.org)

Detection of Pancreatic Duct in Computed Tomography

Volumetric Cerebral Vessel Labeling

Advanced Machine Learning Techniques for Data-Driven Monitoring of Coil Winding Processes in Electric Motor Manufacturing

Numerous industrial processes generate process data that can offer valuable insight into the manufacturing process. These datasets may include different types of information, such as temperature profiles, pressure measurements, and force-time curves, all of which are essential to understanding the dynamic nature of industrial processes. Driven by this, the evaluation of sequential data, specifically time series data, through appropriate machine learning (ML) techniques, particularly deep learning (DL), has gained significant attention. The possible applications of data-driven monitoring utilizing ML encompass detecting anomalies in manufacturing operations, predicting maintenance requirements to minimize downtime, optimizing production parameters for improved efficiency, and determining the resultant product quality.

The winding of coils is a fundamental process step in producing electric motors, generators, electromagnetic actuators, and transformers. The demand for these items, and consequently for coil winding, has increased with the transition to sustainable energy sources. DIN 8580 defines winding as the process of continuously bending the wire around a core part, such as a bobbin or stator tooth, to join them. For specific applications, especially small motors, the application of non-overlapping concentrated windings has become a common practice. The windings are wrapped around one tooth per coil, which increases mass production capabilities and reduces the size of winding heads, in line with the requirements of the automotive industry. The needle winding technique is widely adopted for producing concentrated windings in laminated cores of entire stators. The automated process performs both winding and coil joining. However, monitoring the stochastic process of needle winding presents a challenge due to numerous variables that can impact it.

In the early stages, Bosch, our industrial partner, upgraded certain winding machines to record all process data that could be important for monitoring purposes. This includes force, torque, and position curves, as well as additional tabular data such as wire length and stator tooth height. Initial attempts to predict the resulting quality of coils based on this data have shown basic potential but have been limited by small amounts of data and comparatively simple ML techniques. With the expansion of the database, our data-driven approach to process monitoring shall be further developed by incorporating more advanced ML/DL techniques. Thus far, limited research has been conducted on the implementation of ML or DL techniques for monitoring manufacturing processes using time series data, as the bulk of previous studies concentrated on time series forecasting for predictive maintenance use, rather than time series classification/regression and anomaly detection for the monitoring of manufacturing processes.

Overall, this thesis aims to develop a monitoring technique powered by data and employing advanced ML methods. The primary objective is to detect anomalies in the coil winding process, which will initiate a comprehensive process monitoring. This will enable engineers to optimize the manufacturing process. Anomaly detection serves two crucial purposes: accurately identifying established fault patterns and detecting previously unseen faults that may have evaded end-of-line testing. Detecting fault patterns reliably significantly improves overall product quality. One objective is to analyze the challenges that arise during the needle winding process and the factors that contribute to these difficulties. This involves using the winding machine’s control signals to assess the ongoing winding stages and detect sources of errors. Another objective is to determine how to handle various product types and machine configurations appropriately. A further objective is to select an appropriate ML or DL model by investigating different data preparation techniques. We plan to identify the best ML method through careful analysis and the establishment of a preliminary standard. A combination of models will also be considered. The final objective is to prototypically integrate the model into the ongoing production process and evaluate its effectiveness in a real-time operational setting. Through all this, we aim to demonstrate the potential of a ML/DL technique for data-based monitoring of coil winding procedures in the production of electric motors. This will enhance the quality of the process, provide valuable process insights, and reduce overall costs.