The Generating Super Resolution of Thermal Image based on Deep
Learning

The Generating Super Resolution of Thermal Image based on Deep
Learning

Super resolution, thermal image, deep learning

The need of high resolution to thermal image is very urgent and important. The high resolution of thermal image is able to give
the accurate information on heat distribution map of the objects. The accurate heat distribution maps is able to give accurate
temperature information. This accurate temperature measurement is used for measuring many objects such as electric motors,
engines, human body and so on. These information used for detecting the anomalies of the object in order to find the damaged
parts. The anomalies are considered as damaged parts find in solar panel, agriculture field, the building, bridges and so on.
As the super resolution of thermal image is very important, that the generating of them is a compulsory. Whereas, the camera
for obtaining the super resolution thermal images are very rare, not available at common market. Furthermore, this kind of
device is very expensive too. Therefore not all the user such farmer or technician are able to have them. In order to handle the
problem, the proposed method has the purpose to generate super resolution thermal image economically and easier through
deep learning method. The dataset is taken from solar panel. The results show that the proposed method is able to handle the
low resolution problem of thermal images.

Ismail1
, Yefriadi2
, Yuhefizar3
, Fibriyanti4
, Zulka Hendri5

Politeknik Negeri Padang

29-04-2022

PDF

Indonesia

Text