Annealed Hopfield Neural Network for Recognizing Partially Occluded Objects

Suk-Hun Yoon

Abstract


The need for recognition of partially occluded objects is increasing in the area of computer vision applications. Occlusion causes significant problems in identifying and locating an object. In this paper, an annealed Hopfield network (AHN) is proposed for detecting threat objects in passengers’ check-in baggage. AHN is a deterministic approximation that is based on the hybrid Hopfield network (HHN) and annealing theory. AHN uses boundary features composed of boundary points and corner points which are extracted from input images of threat objects. The critical temperature also is examined to reduce the run time of AHN. Extensive computational experiments have been conducted to compare the performance of the AHNwith that of the HHN.


Full Text:

PDF

References


Agliari, E., Barra, A., and Notarnicola, M., “The relativistic Hopfield network: rigorous results,” Journal of Mathematical Physics, Vol. 60, No. 3, pp. 1-11, 2019.

Aiyer, S. V. B., Niranjan, M., and Fallside, F., “A Theoretical Investigation into the Performance of the Hopfield Model,” IEEE Transactions on Neural Networks, Vol. 1, No. 2, pp. 204-215, 1990.

Cen, F. and Wang, G., “Boosting Occluded Image Classification via Subspace Decomposition-Based Estimation of Deep Features,” IEEE Transactions on Cybernetics, Vol. 50, No. 7, pp. 3409-3422, 2020.

Cong, Y., Tian, D., Feng, Y., Fan, B. and Yu, H., “Speedup 3-D texture-less object recognition against self-occlusion for intelligent manufacturing,” IEEE Transactions on Cybernetics,” Vol. 49, No. 11, pp. 3887-3897, 2019.

de Castro, F. Z. and Valle, M. E., “A broad class of discrete-time hypercomplex-valued Hopfield neural networks,” Neural Networks, Vol. 122, pp. 54-67, 2020.

Hopfield, J. J. and Tank, D. W., “‘Neural’ Computation of Decisions in Optimization Problems,” Biological Cybernetics, Vol. 52, pp. 141-152, 1985.

Hopfield, J. J. and Tank, D. W., “Computing with neural circuits: a model,” Science, Vol. 233, pp. 625-633, 1986.

Hopfield, J. J., “Neural networks and physical systems with emergent collective computational abilities,” Proceedings of the National Academy of Sciences of the United States of America, Vol. 79, pp. 2554-2558, 1982.

Hopfield, J.J ., “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, pp. 3088-3092, 1984.

Kim, J. H., Yoon, S. H., Kim, Y. H., Park, E. H., and Ntuen et al., “Efficient matching algorithm by a hybrid Hopfield network for object recognition,” Proc. SPIE 1709, Applications of Artificial Neural Networks III, Orlando, FL, September 16, 1992.

Kortylewski, A., Liu, Q., Wang, A., Sun, Y., and Yuille, “A., Compositional convolutional neural networks: a robust and interpretable model for object recognition under occlusion,” International Journal of Computer Vision, Vol. 129, pp. 736-760, 2021.

Montgomery, D. C., Design and Analysis of Experiments (10th Ed.), Wiley, New York, 2020.

Nasrabadi, N.M. and Li, W., “Object recognition by a Hopfield neural network,” IEEE Transactions on Systems, Man, and Cybernetics, Vol. 21, No. 6, pp. 1523-1535, 1991.

Priya, L. and Anand, S., “Object recognition and 3D reconstruction of occluded objects using binocular stereo,” Cluster Computing, Vol. 21, pp. 29-38, 2018.

Sohn, K. Alexander, W. E., Kim, J. H., and Snyder, W. E., “A constrained regularization approach to robust corner detection,” IEEE Transactions on System, Man, and Cybernetics, Vol. 24, No. 5, pp. 820-828, 1994.

van den Bout, D. E. and Miller III, T.K ., “Graph partitioning using annealed neural networks,” International 1989 Joint Conference on Neural Networks, Washington DC, USA, pp. 521-528.

Wang, X.-Y. and Li, Z.-M., “A color image encryption algorithm based on Hopfield chaotic neural network,” Optics and Lasers in Engineering, Vol. 115, pp. 107-118, 2019.


Refbacks

  • There are currently no refbacks.