Improving Stereo Matching Quality with Scanline-Based Asynchronous Hopfield Neural Networks

運用非同步式霍普菲爾類神經網路求解掃描線式之雙影像對應問題

10.29977/JCIIE.200701.0005

孫德修(Te-Hsiu Sun)

雙影像對應 ； 霍普菲爾類神經網路 ； 雙影像系統 ； stereo matching ； Hopfield neural network ； scanline ； correspondence problem

工業工程學刊

24卷1期（2007 / 01 / 01）

50 - 59

英文

雙影像對應問題為雙影像系統中之重要研究課題，本文之目的在以非同步式霍普菲爾類神經網路求解掃描線式之雙影像對應問題。研究中首先運用Sobel operator擷取雙影像中物件邊點，再將每一條掃描線中所得之特性點之對應關係轉換成0-1整數規劃問題，其目標函數為一相異性函數，限制式包括視差及單一性；此規劃問題更進而轉換成一能量函數之形式，最後以非同步式霍普菲爾類神經網路求取對應相異性之最小值或其近似值。本研究並以一組常用之雙影像圖例做驗證，實驗結果顯示本研究所提出之方法可應用於PC之平台下，且明顯改善所得對應解之品質。

Stereo matching problem has been a critical task in a stereovision system. This paper presents an asynchronous Hopfield neural network for solving a scanline-based stereo matching problem. The primitive of matching is the edge point extracted using Sobel operator in stereo images. Then, the matching problem is formulated as a '0-1' integer programming with a dissimilarity objective function and disparity and uniqueness constraints, and subsequently is transformed into the form of an energy function scanline by scanline. Finally, the asynchronous Hopfield neural network is used to obtain the minimum energy value that corresponds to the approximated optimal solution of the problem. A set of commonly used stereo images is used to verify the proposed method. Experimental results show that the proposed method can be suitably implemented on the PC platform, and gain efficiency by using scanline-based matching criterion.

1. Ackermann, F.(1984).Digital image correlation- performance and potential application in photogrammetry.Photogrammetric Record,11,429-439.
2. Barnard, S.T.,M. A. Fisher(1982).Computational stereo, Computing Surveys,14,553-572.
3. Calitz, M.-F.,H. Ruether(1985).Least absolute deviation (LAD) image matching,52,160-168.
4. Cavalieri, S.,A. D. Stefano,O. Mirabella(1994).Optimal path determination in a graph by Hopfield neural network.Neural Networks,7,387-404.
5. Foerstner, W.(1982).On the geometric precision of digital correlation.International Archives of Photogrammetry,24,176-189.
6. Giaquinto, N.,M. Savino,S. Taraglio(2002).A CNN- based passive optical range finder for real-time robotic applications.IEEE Transactions on Instrumentation and Measurement,51,314-319.
7. Grimson, W.(1985).Computational experiments with a feature based stereo algorithm.IEEE Transactions on Pattern Analysis and Machine Intelligence,PAMI-7,17-34.
8. Gruen, A.(1985).Adaptive least square correlation: A powerful image matching technique.ISPRS Journal of Photogrammetry, Remote Sensing and Cartography,14,175-187.
9. Gruen, A.,P. Agouris(1994).Linear extraction by least squares template matching constrained by internal forces.Proceedings of ISPRS Commission III Symposium on Spatial Information from Digital Photogrammetry and Computer Vision,509,316-232.
10. Hertz, J.,A. Krogh,R. G. Palmer(1991).Weighted point matching, Introduction to the theory of neural computation.Addison-Wesley publishing company,72-79.
11. Hopfield, J.(1982).Neural networks and physical systems with emergent collective computational abilities.Proceedings of the National Academy of Sciences of the United States of America,79,2554-2558.
12. Hopfield, J.(1982).Neurons with graded response have collective computational properties like those of two- state neurons.Proceedings of the National Academy of Sciences of the United States of America,81,3088-3092.
13. Hopfield, J.,D. W. Tank(1985).Neural` computation of decisions in optimization problems.Biological Cybernetics,52,141-152.
14. Hu, J. E.,P. Siy(1993).An ordering-oriented Hopfield network and its application in stereo vision.SPIE 1965, Applications of Artificial Neural Networks IV,556-567.
15. Lee, J. J.,J. C. Shim,U. H. Ha(1994).Stereo correspondence using the Hopfield neural network of a New energy function.Pattern Recognition,27,1513-1522.
16. Lee, S. H.,J. J. Leou(1994).A dynamic programming approach to line segment matching in stereo vision.Pattern Recognition,27,961-986.
17. Medioni, G.,R. Nevatia(1985).Segment based stereo matching.Computer Vision, Graphics, and Image Processing,31,2-18.
18. Mortara, M.,M. Spagnuolo(2001).Similarity measures for bending polygonal shapes.Computer & Graphics,35,13-27.
19. Mousavi, M. S.,R. J. Schalkoff(1990).A parallel distributed algorithm for feature extraction and disparity analysis of computer images.Proceedings of IEEE Computer Society Conference on Computer Vision on Pattern Recognition,9(13),428-435.
20. Mousavi, M. S.,R. J. Schalkoff(1990).Stereo vision: a neural network application to constraint satisfaction problem.SPIE 1382 Intelligent Robots and Computer Vision IX: Neural, Biological, and 3-D Methods,228-239.
21. Nasrabadi, N.M.,W. Li(1991).Object recognition by a Hopfield neural network.IEEE Transactions on Systems, Man, and Cybernetics,21,1523-1535.
22. Nasrabadi, N.M,C. Y. Choo(1992).Hopfield network for stereo vision correspondence.IEEE Transactions on Neural Networks,3,5-13.
23. Ohta, Y.,T. Kanade(1985).Stereo by intra-and-interscanline search using dynamic programming.IEEE Transactions on Pattern Analysis and Machine Intelligence,PAMI-7,139-154.
24. Ozanian, T.(1995).Approaches for stereo matching-A review.Modeling Identification Control,16,65-94.
25. Pajares, G.,J. M. Cruz(2001).Local stereovision matching through the ADALINE neural networks.Pattern Recognition Letters,22,1457-1473.
26. Pajares, G.,J. M. Cruz,J. Aranda(1997).Relaxation by Hopfield network in Stereo image matching.Pattern Recognition,31,561-574.
27. Parvin, B.,G. Medioni(1991).A layered network for the correspondence of 3D objects.Proceedings of the 1991 IEEE International Conference on Robotics and Automation,1808-1813.
28. Ruicheck, U.,J. G. Postare,J. L. Bruyelle(1998).A neural approach for obstacle detection with a linear stereoscopic sensor.Mathematical and Computer Modelling,27,215-228.
29. Ruicheck, Y.,J. G. Postaire(1995).A neural implementation for high speed processing in linear stereo vision.1995 IEEE International Conference on Systems, Man and Cybernetics 5, Oct.,390-3907.
30. Shirai, Y.(1987).Three-Dimensional Computer Vision.New York:Springer-Verlag.
31. Tank, D.,J. Hopfield(1986).Simple Neural' optimization networks: An A/D converter, signal decision circuit, and a linear programming circuit.IEEE Transactions on Circuits and Systems,CAS-33,533-541.
32. Tien, F.C.,C. A. Chang(1999).Neural network for precise 3D measurement in stereo vision system.International Journal of Production Research,37,1935-1948.
33. Wilson, G. V.,G. S. Pawley(1988).On the stability of the traveling salesman problem Algorithm of Hopfield and Tank.Biological Cybernetics,58,63-70.
34. Wolfe, W. J.(1991).A parallel approach to simultaneously solving the correspondence problem and the pose estimation problem.Proceeding of the SPIE Conference on Mobile Robots,1613,120-126.

1. Sun, Te-Hsiu(2009).STEREO MATCHING USING SYNCHRONOUS HOPFIELD NEURAL NETWORK.工業工程學刊,26(4),276-288.