Mathematical Sciences

<h4>An automated system for detecting stress in farm animals using Sound Event Detection (SED) via machine learning</h4>
<p>Location: MV:L15, Chalmers tvärgata 3</p>
<p>Abstract: Monitoring animal well-being is crucial for optimizing livestock management and ensuring animal welfare. Traditional stress assessment methods, such as visual inspection and physiological measurements, are time consuming and disruptive. This study explores the development of an automated system for detecting stress in farm animals using Sound Event Detection (SED) via machine learning. Various audio classification models were evaluated, focusing on Per-Channel Energy Normalization (PCEN) and wavelet-based feature extraction to improve robustness across different microphone types. The study further investigates Federated Learning (FL) techniques, including Federated Averaging (FedAvg) and Soft Clustered Federated Learning (FedSoft), to enhance model generalization under non-IID data conditions. The results indicate that, while PCEN-based models yield high F-scores, wavelet-based methods demonstrate better generalization across diverse recording setups. Federated learning approaches show promise but require further optimization to address performance inconsistencies. These findings contribute to the development of scalable and adaptable SED models for real-world livestock monitoring applications.</p>

Architecture and Civil Engineering

Bachelor’s thesis Chemistry and Chemical Engineering

Chemistry and Chemical Engineering

Communication and Learning in Science

Computer Science and Engineering

Electrical Engineering

Industrial and Materials Science

Life Sciences

Mechanics and Maritime Sciences

Microtechnology and Nanoscience

Physics

Space, Earth and Environment

Technology Management and Economics

Presentation of master thesis

Choose subject

Course syllabus for

Subject

Each academic year contains different courses

See programme page at universityadmissions.se

Educational area

Permissible substitutions of compulsory courses

To see your programme syllabus – select the academic year in which you started the programme.

[en: page.studyprogrammepage.syllabusesview.coursesdescription]

Presentation of master thesis

Master thesis presentations at Mathematical Sciences

31 March 2025

10:00–11:30 Anton Forsberg