Title :
A Local Image Interpolation Method Based On Gradient Analysis
Author :
Yan, Tian ; Bin, Liu ; Tao, Li
Author_Institution :
Dept. of Electron. & Information Eng., Huazhong Univ. of Sci. & Technol., Wuhan
Abstract :
Image interpolation has been widely used and studied in the fields of image processing. Considering the different complexities along the different directions, a novel image interpolation method based on gradient analysis is presented in this article. First, for each point that need to be interpolated, the gradient values along the horizontal and vertical directions are estimated respectively. Second, different interpolation methods are carried out along these two directions according to the gradient values, that is a higher order interpolation method is applied if the gradient value is greater than certain threshold; otherwise, a lower order interpolation method is used. This method gives an attention to the precision and complexity of interpolation procedure. Experimental results show that the proposed method is feasible and promising
Keywords :
gradient methods; image processing; interpolation; gradient analysis; gradient value estimation; image processing; local image interpolation; Anisotropic magnetoresistance; Image analysis; Image processing; Information analysis; Intelligent control; Interpolation; Laboratories; Low pass filters; Polynomials; Root mean square; Gradient; image interpolation; root mean square error;
Conference_Titel :
Neural Networks and Brain, 2005. ICNN&B '05. International Conference on
Conference_Location :
Beijing
Print_ISBN :
0-7803-9422-4
DOI :
10.1109/ICNNB.2005.1614829