Parameter controlled chaotic synergetic neural network for face recognition

Neural network plays a major role in the field of pattern recognition. For pattern recognition, a major drawback with traditional neural networks is that neural networks may easily be trapped in spurious states. Synergetic neural network (SNN) has been proposed in the literature to overcome this problem, however, when applying synergetic neural network on face recognition, the results are not satisfactory for large image databases due to low memory capacity. Therefore, the chaotic dynamic property is introduced to the conventional synergetic neural network in order to resolve the problem. In this paper, an additional control parameter is introduced to the chaotic synergetic neural network (CSNN) in order to terminate the recognition process whenever an image is recognized. This helps to alleviate processing memory demand which often accompanies such networks. Various imagery defects are tested and the accuracy of both methods is evaluated based on incremental sample size.