Up-scaling Method of Images Based on Fractal Theory
The basic idea behind a multi-scale representation is to embed the original image into such a one-parameter family of derived images, which become more and more smooth and simple. The simplification means the loss of details and information. It is important to measure the information of an image at different scale and kept the spatial and spectral information in the upscaling images. Fractal theory can be used as a quantitative index to comprehensively describe the complicated structural information of images. The pixel fractal model is suitable for analyzing the mico-structure surrounding each pixel in a particular local. In this article, the pixel fractal model was introduced and used to analyze the structural characteristics of images. Based on the information, a upscaling model was presented which consisted of the pixel fractal model. Compared to mean filter and median filter with resolutions of images at different scales, the results showed that 1) mean filter method and the approach presented could both maintained well in the mean and self-relevant characteristics (Moran’I and Geary’s). 2) the upscaling model based on the pixel fractal was superior to mean filter method in preserving structural information 3) median filter method could not reflect the structural feature of the image because of the fast decay of self-relevance.
pizel fractal model upscaling method spatial information
Cao Liqin Li Pingxiang Zhang Liangpei
School of Printing and Packaging Wuhan University, 129 Luoyu Road, Wuhan 430079, P.R.China State Key Laboratory of information Engineering in Surveying, Mapping and Remote Sensing(LIESMARS)Wu State Key Laboratory of information Engineering in Surveying, Mapping and Remote Sensing(LIESMARS)
国际会议
The 31st International Congress on Imaging Science(第31届国际影像科学大会 ICIS2010)
北京
英文
622-625
2010-05-12（万方平台首次上网日期，不代表论文的发表时间）