正态分布模拟倾斜刃边图像的MTF测量方法

MTF Measurement Method Based on Slanted-Edge Image Simulated by Normal Distribution

调制传递函数(MTF)是评价光学成像系统信息传递频率特性的一个关键指标,倾斜刃边法是测量MTF的主要方法之一。为了解决传统倾斜刃边法对噪声图像边缘角度估计的不准确性,提出了一种基于正态分布拟合和中值点查寻的MTF测量新算法。通过正态分布函数模拟一幅与实际图像误差最小的模拟图像,利用中值点查寻法对模拟图像处理,求得最佳刃边角度,利用该角度求出实际图像的边缘扩散曲线,再通过费米函数拟合获得边缘扩散函数(ESF)。由实验结果可以看出,在图像噪声较大的情况下,对倾斜刃边角度估计的准确度得到了明显提升,进而也提高了MTF的测量精度。

The modulation transfer function（MTF） is one of the key indicators to evaluate the signal transmission frequency characteristics of an optical imaging system, and the slanted-edge method is one of the important MTF measurement methods. To solve the problem that the traditional slanted-edge measurement method is not robust against noise, a new MTF measurement method based on the normal distribution fitting and median point estimate is proposed. A modified method that estimates the edge angle by fitting a two-dimensional function to the image data based on the normal distribution is proposed, the optimum slope angle and edge-spread curve of the actual image are calculated by using the method of looking up median point to the fitted image, and the edge-spead function（ESF） is achieved by using the Fermi function. The experimental result shows that the precision of the edge angle of the modified method is improved obviously when the noise of the actual image is relatively seriously,and the precision of the value of the MTF is also increased.