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2 N,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 influence of the sampling window on the measurement of optical system MTF test was studied. In the standard MTF test, using the discrete Fourier transform calculation, the length of the line diffusion function scan direction of the selected pinhole line diffusion function half-width of about 5 times the MTF test results and the design value of the difference between the extreme value of 0.011; When calculated by Fast Fourier Transform, the length of the line spread function is about 5 times of the FWHM of the selected pinhole line diffusion function, and the number of sampling points should satisfy 2 N at the same time. The difference between the test result and the designed value 0.010. In order to prove the universality of this conclusion, according to the principle of selection of sampling window, the MTF of the large axial and axial axes of the large aberration lens was tested. The test results were compared with the OPTIKOS test results, the maximum difference was 0.013. The experimental results show that this conclusion can provide the basis for the sampling window selection in optical system MTF test.