Assessment of Image Quality for Spot-Detecting Camera with Anisotropic Gaussian Fitting

LI Kun,ZHAO Jian-ke,CHEN Yong-quan,XUE Xun,DUAN Ya-xuan,CAO Kun
DOI: https://doi.org/10.3788/ope.20172512.3187
2017-01-01
Abstract:To evaluate quantitatively the image quality of a spot-detecting camera,a 2D anisotropic Gaussian fitting method was proposed.A spot-detecting camera was used to collect the infinity point target coming from a star simulator,and a certain area of energy distribution was obtained.Based on the correlation of the Gaussian energy distribution and the original distribution,this energy distribution was used in anisotropic Gaussian fitting to solve the parameters such as Gaussian radius and anisotropic factor.Then,the size of 80% of the energy was obtained by numerical integration.The experimental results show that the background threshold and anisotropic factor are optimized automatically and the influence of the noise on test results is reduced since the correlation is introduced.Moreover,the introduction of anisotropic factors,evaluation of pixel-phase error and the effect of optical system aberration on the imaging quality of the camera can be used as the testing condition criterion for defocused spots.The experimental results show that the repeatability of equivalent area circle diameter for the defocused spots is 0.15 pixel under the crisscross-shaped condition and that is 0.19 pixel under the farmland-shaped condition,respectively.The proposed method implements the defocused spot test by spot-detecting cameras,and also controls better imaging quality.
What problem does this paper attempt to address?