Interpolation processing method for digital image

Abstract

The present invention provides a kind of interpolation processing method of digital image, at first determine the corresponding image point coordinates (x, y) of the pixel point of the target image in the source image, and then according to the coordinates (m, n) of the adjacent pixel points of the image point ), determine the Gaussian weight W i , and then according to the Gaussian weight W i and the value V of the adjacent pixel i , calculate the value f(x, y) of the image point, and interpolate through Gaussian weights, while reducing the interpolation processing time, and relatively improving the quality of the image.

Description

technical field [0001] The invention relates to a digital image processing method, in particular to a method for interpolating digital images using Gaussian weights. Background technique [0002] When resizing or transforming an image, the original pixels are rearranged and new pixels are generated. When the image is enlarged, the original pixel gap expands, and some new pixels are needed to fill the gap. When the image is reduced, the pixels start to mix and generate new pixels. In order to fill the generated gap, the color value of the adjacent pixel must be calculated and used. This value is used to fill or interpolate gaps. This is what we call interpolation. The interpolation method is the method used when the image redistributes pixels. The purpose is to determine the information of the gap points between known points in the image. [0003] There are many traditional digital image interpolation methods, among which the fastest is NearestNeighbor (nearest neighbor), w...

AI Technical Summary

Problems solved by technology

[0003] There are many traditional digital image interpolation methods, among which the fastest is NearestNeighbor (nearest neighbor), which simply copies adjacent pixels. Although this method is fast, it often produces aliasing; Bilinear The (bilinear interpolation) method takes the average color of each pixel and its four adjacent pixels, and then creates a moderate shadow between pixels to smooth the transition. The effect is better, but it takes a long time; The Bicubic calculation method (bi-cubic interpolation) is similar to Bilinear, the difference is that it calculates the color average value of each pixel and the eight pixels in the direction of the adjacent "m" font. This algorithm works best, but it is the most time consuming

The above interpolation methods all have disadvantages, either the effect is not good, or it takes a long time. Therefore, after the image is transformed, how to complete the image interpolation in a short time to reduce the waiting time for interpolation , is currently in digital image processing, the urgent need for products and the focus of the direction that developers should work hard

Images