改进的三次卷积插值的减点数算法  

Improved Cubic Convolution Interpolation Algorithm with Less Points

作　　者：上官晋太[1] 

机构地区：[1]长治学院计算机系,山西长治046011

出　　处：《山西大学学报（自然科学版）》2015年第1期79-84,共6页Journal of Shanxi University(Natural Science Edition)

基　　金：山西省高校科技开发项目(20111128);山西省青年科技研究基金(2012021015-2);山西省高等学校科技创新项目(2013160)

摘　　要：三次卷积插值算法是图像插值中最常用的算法之一,和线性插值算法相比有较高的插值精度,但其运算量较大,因而插值速度较慢。在二维插值中三次卷积算法需要16个点参与运算。通过利用图像中邻近点之间的相关性使所需点数减少到9个。减少点数后的算法中每次插值需要的乘法次数和加法次数分别减少到三次卷积法的33.33%和40%。新算法在保持和三次卷积算法同样插值精度的情况下减少了运算量,并有效地提高了插值运算的速度。Cubic convolution interpolation algorithm is one of the most widely used image interpolation methods.Compared with the linear interpolation,it has a more accurate result but a lower interpolation speed due to its computational complexity.The cubic convolution algorithm needs sixteen points in two-dimensional image interpolation.The new algorithm proposed reduces the necessary points to nine by using the correlativity of neighbor pixels.Furthermore,the times of multiplication and addition computation in each interpolation can be reduced to 33.33% and 40% of that in original cubic convolution,respectively.The new algorithm has lower computational complexity and higher interpolation speed while keeping the same precision as cubic convolution.

关 键 词：三次卷积插值 减点数三次卷积插值 相关性 插值精度 

分 类 号：TN911.73[电子电信—通信与信息系统]