Title :
A method for single frame super resolution with inpainting based on sparse dictionary learning
Author :
Umehara, Takuya ; Mimura, Kazushi ; Hirabayashi, Akira
Author_Institution :
Grad. Sch. of Inf. Sci., Hiroshima City Univ., Hiroshima, Japan
Abstract :
We discuss a method for single frame super resolution with inpainting using the sparse representation of the image and the dictionary learning. We experimentally show that treating super resolution and inpainting simultaneously decreases the salt and pepper noise than super resolution after inpainting.
Keywords :
image representation; image resolution; image restoration; image inpainting; pepper noise; salt noise; single frame super resolution; sparse dictionary learning; sparse image representation; DH-HEMTs; Dictionaries; Feature extraction; Image resolution; Sparse matrices; Training; Vectors;
Conference_Titel :
Asia-Pacific Signal and Information Processing Association, 2014 Annual Summit and Conference (APSIPA)
Conference_Location :
Siem Reap
DOI :
10.1109/APSIPA.2014.7041734
