Tomasz Barszcz
Papers
6
Total Citations
54
H-Index
4
About
Tomasz Barszcz is a researcher specializing in computer vision, machine learning, and automated condition monitoring systems, with a particular focus on industrial infrastructure in mining environments. His work centers on developing intelligent inspection technologies that reduce the need for labor-intensive manual monitoring of belt conveyor systems — critical components in opencast mining operations. Barszcz has made significant contributions to the application of deep learning for thermal image analysis, pioneering approaches that leverage convolutional neural networks (CNNs) and architectures such as U-Net for detecting and segmenting overheated conveyor idlers — a key indicator of mechanical failure. His 2022 paper on infrared image processing pipelines for robotic inspection has garnered 21 citations, establishing a foundational framework in the field, while subsequent work on binary classification CNNs and semantic segmentation has further refined automated fault detection capabilities. A recurring theme across his research is the deployment of mobile inspection robots in hazardous environments, where he integrates heterogeneous sensor data and ensemble classification methods to enhance diagnostic reliability. His most recent work explores dynamic weighted voting strategies for bearing monitoring, reflecting a growing sophistication in multi-sensor fusion. Collectively accumulating over 50 citations, Barszcz's contributions are shaping the future of intelligent, robot-assisted industrial maintenance.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4
- 5
- 6