Title :
Method of single image super-resolution enhancement based on fractal coding
Author :
He Si-hua ; Wu Zhen
Author_Institution :
Unit 92493, Post-Doctoral Scientific Res. Workstation, PLA, Huludao, China
Abstract :
According to the local characteristics of image similarity and transitivity, self-similarity and scale invariance characteristic of fractal characteristics, use fractal coding as the characteristic of similarity evaluation criteria, search internal image blocks and use the low resolution image analogy fusion in a single image to reconstruct a high resolution image information blocks, implement super-resolution enhancement based on fractal coding. The experimental results show that the fractal coding method can keep the image details better, the reconstruction error is smaller, is better than the bilinear interpolation and bicubic interpolation method.
Keywords :
image coding; image enhancement; image resolution; fractal coding; high resolution image information block reconstruction; image similarity; image transitivity; local characteristics; low resolution image analogy fusion; scale invariance characteristic; search internal image blocks; similarity evaluation criteria; single image super-resolution enhancement; Fractals; Image coding; Image enhancement; Image reconstruction; Image resolution; Interpolation; Reflection; fractal coding; self analogy; single image; super resolution enhancement;
Conference_Titel :
Computer Science and Network Technology (ICCSNT), 2013 3rd International Conference on
Conference_Location :
Dalian
DOI :
10.1109/ICCSNT.2013.6967280