What is it about?

Human perception is only capable of perceiving a few objects outside the range of wavelengths for visible light in the electromagnetic spectrum. It restricts humans' ability to discriminate things in a variety of situations, such as dim light or under smoke and fog. The development of thermographic imaging technology has made it possible to see items that are invisible to the naked eye. This enables its usage in many sectors, such as defence, agriculture, healthcare, etc. Thermal cameras have low spatial resolution in comparison to the same-range RGB cameras due to hardware constraints. A deep neural network architecture, SRDRN, is proposed in this study for Super-Resolution (SR) of IR images. SRDRN uses the channel splitting concept with residual learning for computationally efficient super resolution. The viability of the proposed design is validated by analysing it with the available thermal image datasets.

Featured Image

Why is it important?

The development of thermographic imaging technology has made it possible to see items that are invisible to the naked eye. This enables its usage in many sectors, such as agriculture, healthcare, etc.

Perspectives

Thermal cameras have low spatial resolution in comparison to the same-range RGB cameras due to hardware constraints. A deep neural network architecture, SRDRN, is proposed in this study for Super-Resolution (SR) of IR images.

Nagendar Yamsani
SR University, Warangal

Read the Original

This page is a summary of: SRDRN-IR: A Super Resolution Deep Residual Neural Network for IR Images, July 2023, Institute of Electrical & Electronics Engineers (IEEE),
DOI: 10.1109/aic57670.2023.10263808.
You can read the full text:

Read

Contributors

The following have contributed to this page