MoboView
Loading...
Date
2026
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
The goal of the diploma thesis MoboView is to develop a system
for analyzing and monitoring temperature data using modern
web and artificial intelligence technologies. Thermal camera data
is used to detect potentially critical temperature trends at an early stage and present them in
a clear and structured way. A key focus is the automated classification of temperature time
series.
The system consists of several main components. The Angular frontend provides the user
interface and allows the management and visualization of multiple cameras within a dashboard.
Temperature data is displayed in charts and can be analyzed over time.
The backend, implemented with Node.js and Express, handles user, role, and camera management
and provides a REST API. It uses a hybrid database architecture: MSSQL for relational data
such as users and permissions, and TimescaleDB for efficient storage and querying of time-series
data.
A central part of the system is the component, which is based on an LSTM model. This model
analyzes temperature sequences and classifies them as ’risky’ or ’not risky’. This enables
the detection of abnormal developments that may indicate potential fire hazards.
Additionally, a proxy server is used to access camera streams and to handle technical constraints
such as CORS and authentication. The modular architecture allows all components to work
together efficiently and forms a scalable system for intelligent temperature monitoring.