Research On The Modeling Method Of Feature Binding In Visual Perception

Color and shape are the most basic image features in the visual perception, and feature binding is a focal problem in visual perception research. When someone saw a picture, the brain how to perceive various features in different regions, how to bind these features in the process of visual perception, and how to bind a whole entirety. This is hot topic of researches on the field of psychology and produces some new ideas and theories at present.With the rapid development of computer technology, by analyzing neuron-mechanism of study on color and shape binding of human brain and making use of advanced computer modeling, this paper built up a cognitive model to simulate the cognitive process and neuron-mechanism. This model revealed visual perception process of human brain and promoted understanding and retrieving of image.To solve this problem about color and shape of image features, the main works in this paper are as follows:At first, this paper introduced situation of study on theory and model of binding mechanism across the board; the second, it analyzed formation and characteristics of PCNN model and applications in image processing in depth- -image division and edge detection.This paper analyzed correlations between image division and edge detection and color and shape features of image, and demonstrated accuracy and effectiveness about that PCNN could solve feature binding problem.By analyzing the problems existed in the conditional PCNN, this paper advanced a PCNN model based on strong. In this model, the total intensity in the neighborhood of neuron pulse can achieve feature representation and the intensity of pulse of neuron itself can accomplish feature binding.This model was used in feature binding and could achieve simulation experiments. The experimental results indicated that this model could well accomplish feature representation and binding. It provides some reference value for further research of feature binding in the visual perception.The fMRI(functional Magnetic Resonance Imaging) technology is able to monitor the specific areas of the brain with high temporal and spatial resolution. Brain-image data in the visual perception obtained by fMRI is more objective and scientific. In the end of the paper, it advanced that the research on feature binding in the visual perception is added fMRI technology, and it provides new ideas for feature binding of color and shape in the image processing.