Spiking Cortical Model And Its Applications

Based on the studies of existing neural models, in this paper the spiking cortical model (SCM) is presented and find that time matrix of the model can be recognized as human subjective sense of stimulus intensity.The new model can obtain good results when it is applied to image segmentation, image noise suppression, image enhancement, and texture retrieval. In this paper, it has a general study on the following works:1. The discussion includes the famous models of spike generations by Hodgkin and Huxley. It is dedicated to discussion of Hopfield model and some models related to it. In addition, a common spiking model is integrate-and-fire model and it also discusses some advanced variations of such models.2. It summarizes the local-connected neural networks, such as linking field model, pulse-coupled neural networks, and locally excitatory globally inhibitory oscillator networks. Then, it studies the spiking cortical model and its properties, working mechanism and parameters setting techniques.3. It summarizes some basis techniques of image segmentation, image noise, and image enhancement, and explores how the spiking cortical model is applied to these applications.4. The series of output pulse images of spiking cortical model represents segment, edge and texture features of the original image, and can be calculated based on several efficient measures and formed a sequence as the feature of the original image. It is applied to characterizing texture images by the sequence for invariant texture retrieval. The experimental results show that the retrieval scheme is effective in extracting rotation and scale invariant features.