MTF Automatic Measurement Method Based on Line Support Area

The degradation function of satellite images is a function that changes with time and location, however, due to the limitation of ground facilities, the degradation function measured by commonly methods can only be a fixed value, which does not conform to the time-varying characteristics of the degradation function, therefore, we propose an adaptive MTF measurement method based on line support area. The method automatically identifies ground reference objects and calculate the MTF value in real-time, provides high-precision degradation functions for subsequent restoration. Firstly, we use gradient detection operators to extract points with high contrast, and calculate the gradient direction of these points as the phase direction to construct a “line support area”, then we use optimized linear fitting method to fit the line support area and locate the line edge in order to avoid the phenomenon of edge points not being on a straight line due to grid display issues, secondly, considering the uniform gray distribution on both sides of the linear edge that meets the MTF calculation, we introduce the KURTOSIS value to determine the candidate area, and then calculate the MTF value from candidate area. Finally, we verifies the accuracy of the proposed algorithm by using the three-line array target of ZY3. it can be found that the measurement accuracy of the MTF curve obtained by the adaptive edge extraction method at the Nyquist frequency is 96.7%, while the measurement accuracy of the MTF curve obtained by manual methods at the Nyquist frequency is 62%. It can be seen that manually selected methods are prone to introducing noise, leading to a decrease in measurement accuracy, which is not conducive to the construction of degraded models, The MTF curve obtained through adaptive methods not only better suppresses the influence of random noise, but also has better measurement accuracy than the MTF curve obtained manually.