一种利用采样率转换方法的灰度图像放大算法  

An grayscale image magnification algorithm with sampling rate changed filter

作　　者：贾中云[1] 

机构地区：[1]杭州师范学院信息工程学院,浙江杭州310036

出　　处：《杭州师范学院学报（自然科学版）》2005年第5期360-363,388,共5页Journal of Hangzhou Teachers College(Natural Science)

摘　　要：数字图像放大是图像处理的基本操作之一,传统放大算法是运用各种插值算法在图像的像素点间插值实现图像的放大,插值时会退化图像的高频成分,使放大图像轮廓变得模糊.为此从另一种角度出发,根据数字图像的频率特性,利用数字信号处理中采样率转换滤波的原理,通过零值内插器对图像内插零值,用FIR低通滤波器进行滤波的方法实现灰度图像的放大.由于该算法是根据图像的频率特性进行放大,只要放大后图像的频率特性不变,就能得到比较理想的放大效果.同时该算法为图像放大算法研究提供了一种新的思路.It is a basic operation in digital image processing that a digital image is magnified. The traditional method is to by insert new pixel with various interpolated algorithm to obtain a magnifying image. While high frequency in a digital image degraded, the magnified image will become blurring in borderline. This paper puts forward a new method of image magnification algorithm by sampling rate changed filter based on the character of frequency in a digital image. It is carried out by inserting equality zeroes between pixels, then filtered by a FIR low pass filter. Being based on studying the character of frequency in a digital image, we can obtain a fairly good effect magnified image with this algorithm.

关 键 词：图像放大 采样率转换 滤波 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]