Noise suppression in images and video sequences, image quality assessment

Noise suppression is an important pre-processing step in digital image analysis that helps us improve the quality of such image data that undergoes further processing. Importantly, the quality of input images can substantially affect the performance of the next steps in the processing chain, hence removing noise is of great research importance. Our algorithms include classical filtering and analysis techniques together with deep learning approaches. We strive to fully understand the impact of noise, commonly of different characteristics, on the performance of image and video analysis algorithms. Finally, we develop the algorithms for thorough quality assessment of digital images, video sequences, and color quantization.