Edge-based High Precision Fast Image Interpolation Algorithm And Its Application

Image interpolation is an important technology in digital image processing. Traditional interpolation algorithm is simple and easily realizable. They had the same basic principal, that's to find a input pixel, weight another pixel nearby ,then output the gray-value ,take no account of the others . The difference is the way of getting the pixels around. Traditional interpolation algorithm obtains a good visional effects, but they can not process the pixel's local character of tempestuousness jump, such as edges, textures, they can lead to the edge detail to be fuzzy in the enlargement image.To eliminating the detail's blur and alias, this paper promoted two kinds of novel algorithms based on the edge gradient interpolation optimized algorithm. This algorithm first checked the edges of the image magnified by traditional interpolation algorithms, Then the edge's neighbors are amended by a self-relativity interpolation algorithm based on multi-direction gradient variety. When multi times magnified the image, we take the way of cycle-revise by little ratio magnification. The experiments proved that by this way, we got the finer image in detail than the traditional way. For RGB combined color images, color fusion was processed first, then checked the edges, we could also get the distinct effectLess computational complexity of the algorithm and better quality are important in video and network signal processing. This article promoted another way of fast image interpolation algorithm based on edge-directed max-relativity algorithm using threshold method, the interpolation of pixels was accomplished by double-channel method. By estimating relative information at high resolution, the algorithm analyzed the multiple directions relativity of local neighbors and made full use of six down-sampled neighbors involved in the high-resolution image. Experimental results showed that the algorithm not only decreased the computational complexity but also substantially improved the visual quality the interpolated images.These two interpolation algorithms had instinct effects when processed in several kinds of image such as the natural scenery, the animal image and the character image. A great deal of experiment data and processed images confirmed their feasibility and the validity.