计算二维数字卷积的重叠相加法研究  

Study on the Overlapping Addition Method to Compute Two-dimensional Convolution

作　　者：杨继业[1] 李雨青[1] 

机构地区：[1]宁夏大学数学计算机学院,银川750021

出　　处：《工程数学学报》2012年第5期655-662,共8页Chinese Journal of Engineering Mathematics

基　　金：国家自然科学基金(61001156);宁夏自然科学基金(nz0938);宁夏大学自然科学基金(ndzr09-41;ndzr10-35)~~

摘　　要：卷积的计算在数字图像处理中有重要作用,对两个长度相差很大的一维序列做卷积时,会出现大延时,为避免这种现象,可采用重叠相加法.本文将该方法推广到二维数字卷积的计算中,并详细地给出了实施步骤.数值实验表明重叠相加法在计算二维数字卷积时可有效减少乘法运算次数,节省计算机内存.Efficient computation of convolution is important in digital image processing. But there is over time delay when we evaluate two one-dimensional series with extremely different lengths. Over- lapping addition method can be applied to overcome the disadvantage. In this paper, the proposed method is extended to the two-dimensional case and the implementation process is presented. Finally, the numerical experiment shows that the overlapping addition method can reduce multiplication oper- ations and save memory cost.

关 键 词：二维数字卷积 重叠相加法 DFT 

分 类 号：TN911.7[电子电信—通信与信息系统]