適應式影像還原─以計算智慧為方法

Adaptive Image Restoration-A Computational Intelligence Approach

10.6382/JIM.201207.0475

李俊賢(Chun-Shien Li)；陳玟彣(Wen-Wen Chen)

類神經模糊系統（NFS） ； 粒子群演算法（PSO） ； 遞迴最小平方估計法（RLSE） ； 雜訊消除 ； 計算智慧 ； Neuro-fuzzy system (NFS) ； Particle swarm optimization (PSO) ； Recursive least squares estimator (RLSE) ； Noise canceling ； Computational intelligence

資訊管理學報

19卷3期（2012 / 07 / 01）

475 - 507

繁體中文

影像訊號進行傳輸或轉換的過程中經常會受到其它訊號的干擾，導致影像失真的情形發生。為解決此問題，本研究針對灰階影像失真之問題提出一計算智慧之方法，以Takagi-Sugeno類神經模糊系統為架構，結合粒子群最佳化演算法（PSO）與遞迴最小平方估計法（RLSE）建構出複合學習演算法，並且應用於適應性雜訊消除的問題，最終目標在於使受干擾之影像能夠儘量還原接近原始影像。此外，本研究中亦針對類神經模糊系統之模糊規則數目之增減對於系統效能是否有顯著的影響作探討。本研究以灰階影像的雜訊消除進行實驗，將所提出的複合學習演算法與標準的粒子群演算法進行比較，實驗結果顯示本研究所提出之方法優於標準的粒子群演算法。

Image signal may be interfered by some unknown mechanisms during transmission or transformation, resulting in image corruption that useful information or details in the image can be lost more or less. In the paper, we propose a computational intelligence approach with the famous adaptive noise canceling framework to the problem of gray-level image restoration. The approach is based on the theory of Takagi-Sugeno neuro-fuzzy system (NFS) and the proposed PSO-RLSE hybrid learning method which includes the famous particle swarm optimization (PSO) and the well-known recursive least squares estimator (RLSE) algorithm. With the NFS-based image restoration system, the research target is to restore gray-level images from their corrupted versions as possible as can be. In the study, we investigated different sizes of rule base to see the influence by the amount of fuzzy If-Then rules to the restoration performance, and we also compared the proposed PSO-RLSE learning method to the PSO learning method for restoration performance. Several results are shown in the paper. Through the experimental results, the proposed PSO-RLSE method outperforms the PSO method, in terms of restoration performance and learning convergence.

1. Chen, Y. C. (2006), FLS (Fuzzy Logic System), EE, FCU..
2. Afonso, M.,Joshi, S. D.(2005).A wavelet based scheme for adaptive noise cancelling from images.Proceedings of the 28th General Assembly of the International Union Radio Science
3. Astrom, K. J.,Wittenmark, B.(1984).Computer Controller Systems: Theory and Design.Englewood Cliffs, NJ:Prentice-Hall.
4. Duan, Y.,Gao, Y.,Li, J.(2006).A new adaptive particle swarm optimization algorithm with dynamically changing inertia weight.Intelligent Information Management Syatems and Technologies,2(2),245-255.
5. Eberhart, R. C.,Kennedy, J.(1995).A new optimizer using particle swarm theory.Proceeding of sixth International Symposium on Micro Machine and Human Science
6. Eberhart, R. C.,Shi, Y.(1998).Comparison between Genetic Algorithms and Particle Swarm Optimization.EP '98 Proceedings of the 7th International Conference on Evolutionary Programming VII
7. Goodwin, C.,Sin, K. S.(1984).Adaptive Filtering Prediction and Control.Englewood Cliffs, NJ:Prentice-Hall.
8. Jang, J. S. R.(1991).Fuzzy modeling using generalized neural networks and kalman filter algorithm.Proceedings of Ninth National Conference on Artficial Intelligence
9. Jang, J. S. R.(1993).ANFIS: adaptive-network-based fuzzy inference system.IEEE Transactions on System, Man and Cybernetics,23(3),665-685.
10. Jang, J. S. R.,Sun, C. T.,Mizutani, E.(1996).Neuro-Fuzzy and Soft Computing.USA:Prentice Hall.
11. Jin, F.,Fieguth, P.,Winger, L.,Jernigan, E.(2003).Adaptive wiener filtering of noisy images and image sequences.IEEE International Conference on Image Processing
12. Kennedy, J.(2007).Some issues and practices for particle swarms.IEEE Swarm Intelligence Symposium
13. Kennedy, J.,Eberhart, R. C.(1995).Particle swarm optimization.IEEE International Conference on Neural Networks
14. Kuo, Y. H.,Lee, C. S.,Chen, C. L.(2000).High-stability awfm filter for signal restoration and its hardware design.Fuzzy Sets and Systems,114(2),185-202.
15. Lee, C. C.(1990).Fuzzy logic in control systems: fuzzy logic controller, part I, part II.IEEE Transactions on System, Man and Cybernetics,20,404-435.
16. Lee, C. S.,Kuo, Y. H.,Yu, P. T.(1997).Weighted fuzzy mean filters for image processing.Fuzzy Sets and Systems,89,157-180.
17. Li, C.,Chiang, T. W.(2010).Complex neuro-fuzzy self-learning approach to function approximation.Lecture Notes in Artificial Intelligence,5991,289-299.
18. Mastin, A.(1985).Adaptive filters for digital image noise smoothing: an evaluation.IPPR Conference on Computer Vision, Graphics, and Image Processing,31,103-121.
19. Shi, Y.,Eberhart, R. C.(1998).A modified particle swarm optimizer.IEEE World Congress on Computational Intelligence
20. Strobach, P.(1990).Linear Prediction Theory: A Mathematical Basis for Adaptive Systems.New York:Springer-Verlag.
21. Takagi, T.,Sugeno, M.(1983).Derivation of fuzzy control rules from human operator's control actions.Proceedings of IFAC Symposium Fuzzy Information, Knowledge Representation and Decision Analysis
22. Takagi, T.,Sugeno, M.(1985).Fuzzy identification of systems and its applications to modeling and control.IEEE Transactions on System, Man and Cybernetics,15,116-132.
23. Watrous, R. L.(1991).Learning algorithms for connectionist network: applied gradient methods of nonlinear optimization.Proceedings IEEE International Conference on Neural Networks,619-627.
24. Widrow, B.,Glover, J. R.,McCool, J. M.(1975).Adaptive noise canceling: principles and application.Proceedings of the IEEE,63,1692-1730.
25. Zadeh, L. A.(1965).Fuzzy sets.Information and Control,8,338-353.
26. Zadeh, L. A.(1973).Outline of a new approach to the analysis of complex systems and decision processes.IEEE Transactions on System, Man and Cybernetics,3,28-44.
27. Zheng, Y.,Ma, L.,Zhang, L.,Qian, J.(2003).Empirical study of particle swarm optimizer with an icreasing inertia weight.Evolutionary Computation 2003 (CEC 2003),Canbella, Australia: