Reserch Of Image Processing Based On MCA

This paper aims to conduct multi-scale image processing and research. In sdigital image processing, image separation of effective is self-evident importance of the work.MCA is Synthesis algorithm of BP and mp.In order to achieve that image is with the best form of sparse. The theory of edge detection is a hot and difidult research in China and abroad, with the most basic information.Description,image compression and image segmentation steps necessary fields. Wavelet theory is only developed in recent decades, contains a wealth of theoretical knowledge and application of long-term future, in signal processing, digital TV, image analysis, seismic survey, bio-medical imaging, machine vision, radar, fractal, pattern recognition, etc. The application areas have received good results.Simulation of a non-photorealistic rendering is the simiulationg of artistic styles. Its development begins to be concerned about non-realistic graphics has become one of the focus. However, non-photorealistic rendering as a new field of study, and at present its measure of uncertainty, which makes the research field in many ways still in the continually evolving.Based on MCA image separation technology, in order to realizing the separation of image morphology. And selectively involved in the paper, two typical experiments:hFirstly, use MCA to inpaint the image,and edge detect it using B-spline image edge detec-tion.Canny edge detection and wavelet package operator.Achieve the contrast experiment.You can get the better result through the experiment.Secondly, the NPR pencil simulation experimental results show that the pencil simulation image which is not separated by MCA and the image sepatated by MCA are different processed form, the effect is a significant gap. This paper presents a new method, better preservation of image details.