Wavelet Interpolation And The Snake Model Application In The Magnified Image Clearer

Currently, resolution of a variety of monitor is getting better and better as the rapid development of the technical level. However, in practical applications, it is still difficult to obtain high-resolution quality image. For example, the picture witch is due to transmission constraints, or is exposed to noise interference, its resolution can not be maintained as the original definition. The low-resolution images have been far from being able to meet people’s needs, therefore, an effective and practical method of zooming image is needed urgently.Various zooming methods have been studied in my paper. Interpolation methods have some limitations, because using continuous approach to image edge, it will be have wrong high-frequency component in the image. This will result in varying degrees of jagged edges and fuzzy of the enlarged image, the quality of the enlarged image will be seriously affected. The wavelet transform combined with the traditional interpolation model is used to zoom image in my paper. Because the wavelet has the nature of the multi-resolution analysis features and gradually locally refined. Methods of zooming images based on wavelet transform can alleviate the adverse effects caused by traditional interpolation to a certain extent. However wavelet transform also has some shortcomings. The first is the enlarged image becomes dim, second is the enlarged image has jagged edges and varying degrees fuzzy of regional. The reason of darkened of zoomed images are analyzed in my paper. And in my paper wavelet amplification method has been improved, the issue of wavelet amplification method to make the image gray is resolved.Because of the partial differential equations have their superiority. Using the partial differential equations to correct the pixels of the enlarged image with the interpolation will improve the visual effect of the enlarged image and ease the jagged edges and blurred edges of the image interpolation. After repeated experiments, we can see that using self-snake model post-processing method to eliminate the image’s blurring effect,witch is brought by interpolation methods, is very effective. And it can inhibit the edge blurring of image, even it can enhance the edge sharpness. At the same time, it has been improved on the basis of self-snake model in this paper, its edge function is improved. Compared with the traditional edge stopping function, the new edge stopping function is smaller, so that we can process the edge of image better.Finally, video image is zoomed by bilinear interpolation method and wavelet interpolation method. And it is corrected by self-snake model. At the same time, experimental interface of various zooming methods is design and developed.