对光学系统MTF测试中采样窗口对其测量影响进行了研究。在对标准镜头MTF测试时,采用离散傅里叶变换计算,线扩散函数扫描方向长度为所选针孔像线扩散函数半高宽的5倍左右时,MTF测试结果与设计值差异极值为0.011;采用快速傅里叶变换计算,线扩散函数扫描方向长度为所选针孔像线扩散函数半高宽的5倍左右时,同时采样点数要满足2N,测试结果与设计值差异极值为0.010。为了证明此结论的普适性,按所提的采样窗口选取原则,对大像差镜头轴上和轴外的MTF进行了试验,并将测试结果与OPTIKOS的测试结果进行对比,最大极差为0.013。试验结果表明,此结论能够为光学系统MTF测试时采样窗口选择提供依据。

The accuracy of the optical system modulation transfer function(MTF) measurement is studied by different sampling window sizes. Standard lens are tested. When scanning length of line spread function is same as the 5 times of the line speared function of peak width at half height, the max difference between test results and the academic results is 0.011 by computing with DFT. When scanning length of line spread function is same as the 5 times of the line speared function of peak width at half height and the number of sampling is 2N, the max difference between test results and the academic results is 0.010 by computing with FFT. Aberrated lens are tested for proving that the conclusion is valid. The max difference between test results of aberrated lens and Optikos MTF testing results is 0.013. The test results show that the conclusion is practical to the most optical lens MTF testing.