Title :
Sparse natural image statistics and their applications to colorization and compression
Author :
Balinsky, Alexander ; Mohammad, Nassir
Author_Institution :
Sch. of Math., Cardiff Univ., Cardiff, UK
Abstract :
Statistical analysis of natural luminance and colour images have shown to display non-Gaussian distributions. We review a recent result on the non-linear sparse filter response distribution observed in, and illustrate its application to the problem of colorizing natural gray images. We further utilise elements of both image statistics and colorization for compression of the chromacity channels of image data.
Keywords :
data compression; image coding; image colour analysis; statistical analysis; chromacity channels; colour images; display non-Gaussian distributions; image colorization; image data; image statistics; natural gray images; natural luminance; non-linear sparse filter response distribution; statistical analysis; Compressed sensing; Image coding; Image color analysis; Image reconstruction; Minimization; Optimization; Pixel; L1 optimisation; Natural images; colorization; compression; compressive sensing; filter response; sparse distributions;
Conference_Titel :
Image Processing (ICIP), 2010 17th IEEE International Conference on
Conference_Location :
Hong Kong
Print_ISBN :
978-1-4244-7992-4
Electronic_ISBN :
1522-4880
DOI :
10.1109/ICIP.2010.5652168