Deep Learning in Image Reconstruction by Enigma Pattern

Some details

 

Fixed Pattern Noise Removal from Thermal Images:
The objective of the project was to remove the fixed pattern noise on thermal cameras which are susceptible to both external (environmental) and internal (built-in) conditions.  

Challenges:
The primary concern was to remove the noise from the screen while preserving the actual image, this means that no additional data (hallucinations)

… should appear after noise removal. Additionally, since there are three different types of noises affecting thermal images, each of the noises had to be removed separately. 

Results:

The low-frequency noise was decreased by 80%.
The number of artifacts was decreased by 30%.
The high-frequency noise was reduced by 20%.

Target of achieving threshold of 25 dB achieved, where

 

30 dB is believed as fully denoised picture.

 

Technologies:

 

Keras, TensorFlow