| Title |
Application of infrared thermal cameras in microcontroller-based systems |
| Translation of Title |
Infraraudonųjų spindulių kamerų panaudojimas sistemose su mikrovaldikliais. |
| Authors |
Ahmad, Waqas |
| Full Text |
|
| Pages |
53 |
| Keywords [eng] |
Infrared Thermal Imaging, Thermal Camera, Embedded Systems, STM32N6, ATI320, MLX90640, Image Enhancement, Histogram-Based Enhancement, Dynamic Range Compression, Percentile Clipping, Thermal Monitoring, Fault Detection, PCB Thermal Analysis, Real-Time Processing, Temperature Measurement, Microcontroller Systems. |
| Abstract [eng] |
Infrared thermography is non contact temperature measurement technique that enables visualization of heat distribution by detecting infrared radiation emitted from objects. In recent years, the integration of thermal sensors with embedded systems has enabled the development of compact and cost-effective thermal imaging solutions. In this thesis which presents the design, implementation, and optimization of a microcontroller based thermal imaging system developed across multiple stages. In the initial phase, a low-resolution MLX90640 infrared sensor was interfaced with an ESP32 microcontroller to acquire thermal data and transmit it wirelessly through a web-based interface. This stage primarily focused on data acquisition, calibration, and real-time visualization. It was extended to next phase for practical applications in electronic circuit analysis. Thermal imaging was used to study the behavior of printed circuit boards (PCBs) under different operating conditions, including fault detection scenarios. The results showed ability of system to detect hotspots and identify abnormal thermal behavior. In the final stage of the project, the system was upgraded to more advanced platform using the STM32N6 Nucleo-144 microcontroller together with the ATI320 thermal camera. As compared to the earlier MLX90640-based setup, ATI320 provides high resolution RAW14 thermal image data due to this image processing was required before the thermal data could be displayed properly. Several processing methods were implemented to improve image visualization, including linear compression, histogram based image enhancement, percentile clipping and dynamic range compression. During testing, these techniques helped improve image contrast and reduced the effect of extremely hot objects dominating the scene. |
| Dissertation Institution |
Vilniaus universitetas. |
| Type |
Master thesis |
| Language |
English |
| Publication date |
2026 |
