Test object image preprocessing for modulation transfer function calculation in optical system testing

Subject of study . The study is focused on preprocessing methods for slit and pinhole test object images used in modulation transfer function (MTF) calculation. This paper presents an approach that accounts for the finite size of the test object in MTF evaluations. The influence of background suppression, detector noise reduction, and scattering spot center determination methods on the resulting MTF error is analyzed. Aim of study. The aim of this study is to develop data preprocessing procedures, including background and noise suppression, image cropping, and scattering spot center localization, and to calculate the MTF from test object images for a hardware–software complex intended for automated optical system testing. Methods. The preprocessing methods are analyzed using test object images acquired on an optical test bench with two different lenses. The MTF calculation error obtained using various preprocessing techniques is evaluated as the maximum absolute deviation from a reference MTF curve. The reference curve is obtained either by an alternative testing method or from theoretical calculations performed in the design stage of the tested lens. Main results. This error analysis shows that background correction has the greatest impact on the calculation results and that its effectiveness is strongly influenced by the noise suppression method. A median filter combined with a repetitive background subtraction technique is proposed to suppress random noise and background illumination. In addition, because the finite size of the test object must be considered, we propose that the MTF can be corrected by dividing it by the spatial spectrum of the object. Accurate determination of the scattering spot center is also shown to be critical, and the quadratic center-of-mass method, which provides the minimum localization error, is recommended. Practical significance. Application of the proposed methods during the assembly and acceptance testing stages of optical instruments will improve the accuracy of quality assessments based on test object images. Concurrently, the complexity of the optical testing procedure is reduced, and the time required for optical system certification is shortened.