- 
352 
Fig 5 MTF estimation results using knife-edge method 
Fig 8 Error analysis result 
It is clear that MTF estimation from three different targets 
obtain relatively consistent results. However, site 3 obtains 
lower MTF and the most error at Nyquist frequency is about 
0.018, which may result from lower contrast and bigger slope. 
7.2 Pulse method target and result 
The target used for pulse method is a dam between different 
fish ponds in the same scene as knife-edge method, it is shown 
in figure 6 and MTF estimation result is shown in figure 7. 
Fig 6 Pulse target for SPOT/Pan MTF estimation 
Fig 7 MTF estimation result using pulse method 
Because the input pulse is about 1.5 pixels, the first 
zero-crossing is located at normalized frequency of 0.6, which 
leads to the bad result aty^ v = o.6 • 
7.3 Analysis 
Figure 8 shows the relative difference between knife-edge and 
pulse method. The biggest error reaches to 0.085, but it is cause 
by zero-crossing. In other frequency domain, the biggest error 
is about 0.04, which indicate that the accuracy of MTF 
estimation methods is quite reasonable. 
8 CONCLUSION 
Modulation transfer function (MTF) is a key measurement of 
imaging systems’ geometric performance. It is not only an 
indicate the performance of imaging system, but also make a 
deconvolution filter whose purpose is to enhance image 
contrast by ground processing. 
This paper systematically introduces the principle of several 
widely-used on-orbit MTF estimation methods (including point 
source/array method, knife-edge method; pulse method, and 
bi-resolution method), their target deployment/selection 
standards, data processing steps, and their advantages and 
drawbacks. We should choose the optimal method according 
the spatial resolution of sensor and target condition. We hope to 
accelerate the construction of remote sensing data quality 
analysis and assessment system with self-dominated 
intellectual property right through the studies and 
implementation of on-orbit MTF estimation. 
REFERENCE 
[1] D. Leger, F. Viallefont, P. Deliot, et al. On-orbit MTF 
assessment of satellite cameras, Post-launch calibration of 
satellite sensors: P.67-76, Taylor & Francis Group, 2004. 
[2] Schowengerdt R. A, Archwamety C, Wrigley R. C. 
“Landsat Thematic Mapper image-derived MTF”, 
Photogrammetric Engineering & Remote Sensing, Vol.51, 
No.9: P.1395-1406, 1985. 
[3] Dennis L. Helder, Taeyoung Choi. "IKONOS satellite on 
orbit modulation transfer function (MTF) measurement 
using edge and pulse Method", South Dakota State 
University, 2002. 
[4] Manjunath Kempaiah Rangaswamy, “Quickbird II 
two-dimensional on-orbit modulation transfer function 
analysis using convex mirror array”, South Dakota State 
University, 2003. 
[5] D.Helder, T.Choi, M.Rangaswamy. In-flight 
characterization of spatial quality using point spread 
function, Post-launch calibration of satellite sensors: 
P.151-170, Taylor & Francis Group, 2004. 
[6] Robinet F, Leger D, Cerbelaud H, et al. “Obtaining the 
MTF of a CCD imaging system using an array of point 
sources: evaluation of performances”, In: IGARSS 1991: 
P.1357-1361, 1991.