If you want to sign up for this seminar, you need to register with the following information in the form

• surname, first name,
• e-mail address,
• matriculation number,

Please note that the registration period starts 03.10.2025 at 08:00 h and ends 24.10.2025 at 23:59 h. All applications before and after this registration period will not be taken into account.

Registration will be possible within the before mentioned time by sending an e-mail with the desired seminar topic, name and matriculation number to This email address is being protected from spambots. You need JavaScript enabled to view it..

Only one student per topic is permitted (first come - first serve).

The registration is binding. A deregistration is only possible by sending an e-mail with your name and matriculation number to This email address is being protected from spambots. You need JavaScript enabled to view it. until Sunday, 27.10.2024 at 23:59 h. All later cancellations of registration will be considered as having failed the seminar.

Students write a scientific report on a topic closely related to the current research of the DSS group.Therefore, potential topics include pattern recognition and machine learning related aspects.

Students will also present their findings in front of the other participants and the DSS group.

The scarcity of labelled data poses a significant challenge in developing effective deep learning models for classifying underwater acoustic signals. The shortage of large, publicly available datasets stems from the high cost of data collection, the confidentiality of acoustic recordings and the difficulty of accurately annotating data. Transfer learning is a technique in which knowledge from a model trained on a large source task (e.g. image recognition) is applied to a different, yet related, target task. It offers a promising approach to mitigating this issue. The aim of this seminar paper is to examine the transfer learning technique, comparing it with existing NN models to explore how it improves the performance of underwater radiated noise classification.

The assessment of speech quality and intelligibility is of great interest, both in speech therapy and in the evaluation of algorithms for improving speech signals with regard to interference factors such as noise and reverberation. The aim of this seminar paper is to summarize and compare various objective assessment methods. In addition to traditional assessment criteria such as STOI, PESQ, etc., approaches based on neural networks should also be used for the comparison.

The monitoring and identification of objects and other items are important tasks of a ship on the high seas. The detection of other ships is particularly important. When a ship moves, it generates characteristic sound waves that depend on the speed, the hull and, above all, the propeller and its speed. These sound waves can be detected with passive SONAR systems. The decision as to whether it is a ship and in which category it can be classified is mainly made by a human SONAR operator. However, this decision can be subjective, so an automated decision process can be used to provide an objective decision and support. For this, machine learning concepts can be used to automatically classify the vessels. Concepts such as Convolutional Neural Networks (CNN), which are able to efficiently extract features from underwater recordings of vessels, are widely used. Both underwater recordings from databases and artificially generated sounds from ship propellers can be used for this purpose.

Direction of arrival (DOA) estimation is a well-known topic in signal processing and is used in various applications. Some examples include target detection in sonar systems or improving the SNR of wireless communication. There are many commonly used techniques, such as beamforming, music algorithm etc. for DOA estimation. Your goal is to review the most widely used machine learning models for DOA estimation and compare them with traditional algorithms in terms of efficiency, complexity and computational effort.

Hand gesture recognition plays a key role in human-machine interaction applications. The hands are equipped with various sensors for this purpose. Different sensors, such as acceleration sensors, magnetic sensors and optical sensors, are used to reconstruct the movement of the hand. The gestures correspond to defined movement patterns. This problem is well suited for various machine learning approaches. The algorithms are specially adapted to the problem and in some cases manage to map the kinematic characteristics. The aim of this seminar paper is to understand the individual approaches, compare them with each other and present them clearly.

Speech is subject to a variety of interference when transmitted over the underwater channel. In addition to interference signals (ships, sea creatures, waves, noise, etc.), one of the interferences that has the greatest impact on the intelligibility of speech is reverberation. This is caused by reflections from various structures, as well as the water surface and ground. The aim of the seminar is to deal with machine learning methods for dereverberation and to weigh up promising approaches against each other. The approaches presented should show a clear connection to underwater signal transmission.

In order to monitor a predefined area, such as a harbor, with a SONAR system, several sub-SONAR systems - so-called nodes - can be used. Each individual node generates a surveillance image of its immediate surroundings, which can then be inserted into a larger overall image. Such a network is also called a multistatic SONAR network. The nodes each have an individual capacity (e.g. a radius R) to monitor an area around them and can move. Now, with the help of these same nodes, a defined area is to be optimally monitored. In this seminar, you will examine how the given problem can be modeled under simplified conditions and how it could be solved using machine learning methods.