论文信息 - New Optimal Tracking Algorithms for Reference with Random Textures

New Optimal Tracking Algorithms for Reference with Random Textures

The detection and location of a target in a scene is a classical problem, pervasive to many image processing applications. When the target appears on a low contrast background and the image is corrupted with additive gaussian noise, the optimal method is the matched filter [1]. However, the matched filter, as well as the improved linear filtering techniques [2] have been shown to perform poorly on many real-world images [3], because such images often do not belong to the class for which linear filtering is optimal.

Mireille Guillaume | Philippe Réfrégier | François Goudail | Thierry Gaidon

[1] B Javidi,et al. Nonlinear joint-transform correlation: an optimal solution for adaptive image discrimination and input noise robustness. , 1994, Optics letters.

[2] F Goudail,et al. Influence of nonoverlapping noise on regularized linear filters for pattern recognition. , 1995, Optics letters.

[3] Donald Geman,et al. Stochastic Relaxation, Gibbs Distributions, and the Bayesian Restoration of Images , 1984, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[4] F Goudail,et al. Optimal detection of a target with random gray levels on a spatially disjoint background noise. , 1996, Optics letters.

[5] Bahram Javidi,et al. Minimum mean-square-error filter for pattern recognition with spatially disjoint signal and scene noise , 1993 .

[6] B Javidi,et al. Optimum receiver design for pattern recognition with nonoverlapping target and scene noise. , 1993, Optics letters.

[7] Charles D. Hendrix,et al. Tradeoffs in the design of correlation filters , 1992, Other Conferences.

[8] Philippe Réfrégier,et al. Optimal location of random targets in random background: random Markov fields modelization , 1996 .

[9] P. Réfrégier. Filter design for optical pattern recognition: multicriteria optimization approach. , 1990, Optics letters.

[10] B Javidi,et al. Limitation of the classic definition of the correlation signal-to-noise ratio in optical pattern recognition with disjoint signal and scene noise. , 1992, Applied optics.