A surface-based approach to 3-D object recognition using a mean field annealing neural network

Abstract Three-dimensional (3-D) object recognition identifies objects in an input image using a modelbase. We present a 3-D object recognition system, in which a symbolic description of the object is generated from the input range data, in terms of the visible surface patches. The segmented surface representation of an object is regarded as a graph, in which nodes contain the information about the individual surface patches whereas links represent the relationships between them such as connectivity. Object recognition is achieved by matching the object graph of a scene with the model graph of a modelbase. In the proposed approach, a mean field annealing (MFA) neural network (NN) is employed as a constraint satisfaction network, in which an energy function robust to occlusion is proposed for effective matching. The energy level is at its minimum when the optimal match is achieved. Simulation results with synthetic and real range images having various occlusion rates are presented to show the effectiveness of the proposed algorithm.

[1]  John J. Hopfield,et al.  Simple 'neural' optimization networks: An A/D converter, signal decision circuit, and a linear programming circuit , 1986 .

[2]  Hon-Son Don,et al.  3-D Moment Forms: Their Construction and Application to Object Identification and Positioning , 1989, IEEE Trans. Pattern Anal. Mach. Intell..

[3]  Steven Gold,et al.  A Graduated Assignment Algorithm for Graph Matching , 1996, IEEE Trans. Pattern Anal. Mach. Intell..

[4]  A. P. Davignon.  Detecting orientation discontinuities in range images by use of directional derivatives , 1992, Proceedings., 11th IAPR International Conference on Pattern Recognition. Vol. III. Conference C: Image, Speech and Signal Analysis,.

[5]  C. D. Gelatt,et al.  Optimization by Simulated Annealing , 1983, Science.

[6]  Wei-Chung Lin,et al.  A hierarchical multiple-view approach to three-dimensional object recognition , 1991, IEEE Trans. Neural Networks.

[7]  Yoshiaki Shirai,et al.  A scene description method using three-dimensional information , 1979, Pattern Recognit..

[8]  Yoshiaki Shirai,et al.  Object Recognition Using Three-Dimensional Information , 1981, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[9]  Andrew W. Fitzgibbon,et al.  An Experimental Comparison of Range Image Segmentation Algorithms , 1996, IEEE Trans. Pattern Anal. Mach. Intell..

[10]  Rae-Hong Park,et al.  Robust Adaptive Segmentation of Range Images , 1998, IEEE Trans. Pattern Anal. Mach. Intell..

[11]  Richard J. Prokop,et al.  A survey of moment-based techniques for unoccluded object representation and recognition , 1992, CVGIP Graph. Model. Image Process..

[12]  Rae-Hong Park,et al.  3D object recognition in range images using hidden markov models and neural networks , 1999, Pattern Recognit..

[13]  Rae-Hong Park,et al.  Edge-based segmentation of range images: integrating roof and step edges , 1996 .

[14]  Nasser M. Nasrabadi,et al.  Object recognition by a Hopfield neural network , 1991, IEEE Trans. Syst. Man Cybern..

[15]  David E. van den Bout,et al.  Graph partitioning using annealed neural networks , 1990, International 1989 Joint Conference on Neural Networks.

[16]  Eam Khwang Teoh,et al.  Hopfield network with constraint parameter adaptation for overlapped shape recognition , 1999, IEEE Trans. Neural Networks.

[17]  Ramakant Nevatia,et al.  Recognizing 3-D Objects Using Surface Descriptions , 1989, IEEE Trans. Pattern Anal. Mach. Intell..

[18]  Heinrich Niemann,et al.  Object recognition using hidden Markov models , 1994 .

[19]  Eric Mjolsness,et al.  A Lagrangian relaxation network for graph matching , 1996, IEEE Trans. Neural Networks.

[20]  Wei-Chung Lin,et al.  A Neural Network Approach to CSG-Based 3-D Object Recognition , 1994, IEEE Trans. Pattern Anal. Mach. Intell..