Enhanced Remote Sensing Imaging in Uncertain Environment Fusing Adaptive Least Squares and Variational Analysis Methods
- Resource Type
- Authors
- José Tuxpan Vargas; Stewart René Santos Arce
- Source
- 2009 Electronics, Robotics and Automotive Mechanics Conference (CERMA).
- Subject
- Bayes' theorem
Pixel
business.industry
Principle of maximum entropy
Probabilistic logic
Entropy (information theory)
Pattern recognition
Artificial intelligence
Iterative reconstruction
Variational analysis
business
Image restoration
Mathematics
- Language
This paper proposes a new approach for the reconstruction of the RS images degraded by composite noise (additive and multiplicative), taking into account the limitations of the sensor system. Our proposal takes advantage of the statistical and probabilistic qualities of the employed Weighted Constrained Least Squares (WCLS) and Robust Bayes Minimum Risk (RBMR) algorithms. The restoration of the image optimized by aggregating the WCLS, RBMR, the maximum entropy and variational analysis techniques.