Abstract In this paper, we propose a new approach of dense disparity map computing based on the neural network from pair of stereo images. Our approach divides the disparit y map computing into two main steps. The first one deals with computing the initial disparity map using a neuronal method B ack(Propagation (BP). The BP network, using differe ntial features as input training data can learn the functional relationship between differential features and the matching deg ree. Whereas, the second one presents a very simple and fast method to refin e the initial disparity map by using image segmenta tion so an accurate result can be acquired. Experimental results on rea l data sets were conducted for evaluating the neural model proposed.

References

[1] Alvarez L., Deriche R., Sanchez J., and Weickert J., Dense Disparity Map Estimation Respecting Image Discontinuities: a PDE and Scalespace Based Approach, Journal of Visual Communication Image Represent , vol. 13, no. 1; 2, pp. 3;21, 2002 .

[2] Baha N. and Larabi S., Obstacle Detection from Uncalibrated Cameras, in Proceedings of Panhellenic Conference on Informatics , Samos, pp. 152;156, 2008.

[3] Bartoli A. and Sturm P., Non Linear Estimation of the Fundamental Matrix with Minimal Parameters, IEEE Transaction on Pattern Analysis and Machine Intelligence, vol. 26, no. 4, pp. 426;432, 2004.

[4] Binaghi E., Gallo I., Fornasier C., and Raspanti M., Growing Aggregation Algorithm for Dense Two;Frame Stereo Correspondence, in Proceedings of 1 st International Conference on Computer Vision Theory and Application, Germany, pp. 326;332, 2006.

[5] Bobik A. and Intille S., Large Occlusion Stereo, International Journal on Computer Vision , vol. 33, no. 3, pp. 181;200, 1999.

[6] Cochran S. and Medioni G., 3;D Surface Description from Binocular Stereo, IEEE Transactions on Pattern Analysis and Machine Intelligence , vol. 14, no. 10, pp. 981;994, 1992.

[7] Fehn C., Cooke E., Schreer O., and Kauff P., 3D Analysis and Imagebased Rendering for Immersive TV Applications, Signal Processing Image Communication , vol. 17, no. 9, pp. 705; 715, 2002.

[8] Feiner S., MacIntyre B., and Seligmann D., Knowledge;Based Augmented Reality, Communication ACM , vol. 36, no. 7, pp. 53;62, 1993.

[9] Gonzalez R. and Woods R., Digital Image Processing , Addison;Wesley Reading, USA, 1992.

[10] Goulermas J. and Liatsis P., Hybrid Symbiotic Genetic Optimisation for Robust Edge;Based Stereo Correspondence, Pattern Recognition, vol. 34, no. 12, pp. 2477;2496, 2001.

[11] Grimson W., Computational Experiments with a Feature Based Stereo Algorithm, IEEE Transactions on Pattern Analysis and Machine Intelligence , vol. 7, no. 1, pp. 17;34, 1985.

[12] Harris C. and Stephens M., A Combined Corner and Edge Detector, in Proceedings of 4 th Alvey Vision Conference Manchester , UK, pp. 147;151, 1988.

[13] Hartley R., Theory and Practice of Projective Rectification, International Journal on Computer Vision , vol. 35, no. 2, pp. 115;127, 1999.

[14] Haykin S., Neural Network: A Comprehensive Foundation , Pentice Hall, USA, 1998.

[15] Hecht R., Counter Propagation Networks, IEEE International Conference on Neural Networks , vol. 26, no. 23, pp. 4979;4983, 1987.

[16] Iddan J. and Yahav G., 3D Imaging in the Studio and Elsewhere, in Proceedings of SPIE 3D Image Capture and Applications IV , Israel, pp. 48;55, 1994.

[17] Kanade T., Yoshida A., Oda K., Kano H., and Tanaka M., A Stereo Machine for Video;Rate Dense Depth Mapping and its New Applications, in Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition , USA, pp. 196;202, 1996.

[18] Kanade T. and Okutomi M., A Stereo Matching Algorithm with An Adaptive Window: Theory 224 The International Arab Journal of Information Techn ology, Vol. 9, No. 3, May 2012 and Experiment, IEEE Transaction Pattern Analysis and Machine Intelligence , vol. 16, no. 9, pp. 920;932, 1994.

[19] Kumar S. and Chatterji B., Stereo Matching Algorithms Based on Fuzzy Approach, International Journal of Pattern Recognition and Artificial Intelligence , vol. 16, no. 7, pp. 883; 899, 2002.

[20] Levenberg K., A Method for Solution of Certain Non;Linear Problems in Least Squares, Quarterly Journal of applied Mathematics , vol. 11, no. 2, pp. 164;168, 1944.

[21] Luong Q. and Faugeras O., The Fundamental Martix: Theory, Algorithms, and Stability Analysis, International Journal of Computer Vision , vol. 17, no. 1, pp. 43;75, 1996.

[22] Miled W. and Pesquet J., Dense Disparity Estimation from Stereo Images, in Proceedings of 3 rd International Symposium on Image/Video Communication, Tunisie, 2006.

[23] Mostefaoui S., Emergence in Collective Robotics: A Case Study, The International Arab Journal of Information Technology , vol. 1, no. 0, pp. 47;53, 2003.

[24] Nasrabadi N. and Choo C., Hopfield Network for Stereo Vision Correspondence, IEEE Transactions Neural Networks , vol. 3, no. 1, pp. 5;13, 1992.

[25] Nevatia R. and Babu K., Linear Feature Extraction and Detection, Computer Graphics Image Processing , vol. 13, no. 1, pp. 257;269, 1980.

[26] Resk B., Berany P., and Hashimoto H., Camera Control with Disparity Matching Stereo Vision by Artificial Neural Networks, in Proceedings of WISES, US, pp. 139;150, 2003.

[27] Scharstein D. and Szeliski R., A Taxonomy and Evaluation of Dense Two;Frame Stereo Correspondence Algorithms, International Journal on Computer Vision , vol. 47, no. 1;3, pp. 7;42, 2002.

[28] Simond N., Obstacle Detection from IPM and Super;Homography, in Proceedings of IEEE IROS International Conference on Robot and System , CA, pp. 4283;4288, 2007.

[29] Zhang Z. and Loop C., Computing Rectifying Homographies for Stereo Vision, in Proceedings of IEEE Conference on Computer Vision and Pattern Recognition , USA, pp. 125; 131, 1999.

[30] Zhong L., Methodes Robustes en Vision: Application Aux Appariements Visuels , Th se de Doctorat de L INP de Grenoble, 1997.

[31] Tardif P. and Zaccarin A., Segmentation D images De Textures Par Des Modeles Multiresolutions , Univerit Laval, 2000.

[32] Middlebury, available at: http:// vision.middlebury.edu/stereo, last visited 2007. Nadia Baha received the Magister in computer science at CDTA in 1991. She is currently a PhD student and a researcher at the Computer Science Institute of University of Sciences and Technologies (USTHB), Algeria. She is an author of numerous publications for conferences, proceedings. Her research interest s include computer vision and mobile robot navigation . Slimane Larabi received the PhD in computer science from the National Institute Polytechnic of Toulouse, France, in 1991. He is currently Professor at the computer Science Institute of USTHB University, where he conducts research on Computer Vision. His interest topics are Image description, Human action recognition, Head pose estimation, knowledge modelling and representation for computer vision.