We explore the use of Cramér–Rao bound calculations for predicting fundamental limits on the accuracy with which target characteristics can be determined by using imaging sensors. In particular, estimation of satellite orientation from high-resolution sensors is examined. The analysis role that such bounds provide for sensor/experiment design, operation, and upgrade is discussed. Emphasis is placed on the importance of including all relevant target/sensor uncertainties in the analysis. Computer simulations are performed that illustrate that uncertainties in target features (e.g., shape, reflectance, and relative orientation) have a significant impact on the bounds and provide considerable insight as to how details of the three-dimensional target structure may influence the estimation process. The simulations also address the impact that a priori information has on the bounds.
© 2003 Optical Society of America
[Optical Society of America ]
(100.2960) Image processing : Image analysis
(100.3190) Image processing : Inverse problems
(100.5010) Image processing : Pattern recognition
(110.3000) Imaging systems : Image quality assessment
David R. Gerwe, Jennifer L. Hill, and Paul S. Idell, "Cramér–Rao analysis of orientation estimation: influence of target model uncertainties," J. Opt. Soc. Am. A 20, 817-826 (2003)