Title :
Novel image zooming method based on sparse decomposition
Author :
Xuan Zhu ; Qiuju Li ; Xufeng Zhang ; Ning Wang
Author_Institution :
Sch. of Inf. Sci. & Technol., Northwest Univ., Xi´an, China
Abstract :
Basing on the fact that the cartoon and texture in one image have different morphological characteristics, we propose a new model to zoom image. Combining with RDWT, the dictionary sparsely representing the cartoon, and WAT, the dictionary sparsely representing the texture, the presented model can effectively obtain the cartoon and texture. Then, we zoom the cartoon by C&E model and zoom the texture by bicubic interpolation. Superpose the zoomed cartoon and texture, the zoomed image will be obtained. As the experiment results show, the new model has good performs for enhancing edge, protecting small curvature and large gradient, and ensuring the texture clear and completion.
Keywords :
image texture; interpolation; RDWT; bicubic interpolation; cartoon; image texture; image zooming method; morphological characteristics; sparse decomposition; Dictionaries; Discrete wavelet transforms; Image edge detection; Interpolation; Mathematical model; TV; C&E model; Image zooming; RDWT; Sparse representation; WAT;
Conference_Titel :
Information and Automation (ICIA), 2014 IEEE International Conference on
Conference_Location :
Hailar
DOI :
10.1109/ICInfA.2014.6932620