Image Zooming Via Constrained Surface Fitting With Edge Constraint

Image resizing is a basic problem in the fields such as image processing, visualization, computer graphics, virtual reality, and so on. Zooming the size of the given image through image resizing and showing a good visual effect play a critical role in the performance of simulation, computer animation, remote sensing, medical imaging and other practical applications.The conventional image interpolation algorithm and the implicit image interpolation algorithm are two major technologies for image zooming. The common used methods like bilinear image interpolation algorithm,cubic image interpolation algorithm, cubic B-spline image interpolation algorithm, and so on, are belong to the conventional image interpolation algorithm. These methods treat the image as Continuous data, and the kernel function of these methods are inflexible in the space without self-adaption. So saw tooth and fuzzy are emerged in the image with high resolution zoomed by these methods. In order to make up for the inadequacy of conventional image interpolation algorithm, self-adaptive image interpolation algorithm has been proposed. These methods use the edge information of the low-resolution image to realize the interpolation of the unknown pixels in the high-resolution image. The explicit adaptive interpolation algorithm is easily affected by the problem such as Image fuzzy and noise and so on. The implicit adaptive interpolation algorithm can be a good solution to the lack of explicit interpolation algorithm, and statistics for local information can implicitly contain the edge information of low-resolution image, which can get unknown pixels adaptively to generate high-resolution image.In this paper we present a new image zooming algorithm based on surface fitting with edge constraint. In surface fitting, we consider not only the relationship of corresponding pixels between the original image and the enlarged image, but also the neighbor pixels in the enlarged image according to the local structure of original image. Furthermore, during surface fitting, more interpolation constraints are used in the new algorithm for improving the precision of the super sampling pixels. The experimental results show that the new method outperforms the previous methods which based on surface fitting.