"Software Acceleration of Video Noise Filtering and its Integration into Real-Time Video Applications

The project is about removing noise from video signals (for example, those taken by a mobile phones or professional cinema camera). Our current software first detects the type and power of the noise and based on that remove the noise from video. This proposal has three objectives. The first is to handle a specific type of noise called structured noise, not handled yet in our denoising software developed in previous MITACS projects. To handle this, both estimator and denoiser should be adapted to this type of noise by first studying this type of noise and then change the parameters (or structure) of our denoiser to have the maximum performance under these conditions. The second objective is to increase the speed of our algorithms to achieve faster processing for time-constrained applications such as TV or cinema. We plan to revisit the structure of our techniques and carefully study how to change our algorithms to achieve higher speed while maintaining denoising quality. Our third objective is to test our techniques in very time constrained applications such as smart phones and investigate ways to reduce complexity and increase speed so to achieve real time processing.

Intern: 
Meisam Rakhshanfar
Superviseur universitaire: 
Aishy Amer
Project Year: 
2014
Province: 
Quebec
Programme: