A model of primary visual sense with wavelets characterization

In this paper, a 3-band wavelets transform, which has two main characteristics of primary visual sense: Gabor function model and 3 scale transform, is used to simulate the processing of simple cell in the visual cortex. Because of higher order statistics properties of the environment, the input response under the multiresolution analysis is decomposed linearly by using the independent component analysis (ICA) and the basis functions at each resolution are extracted. The experimental results indicate that our model with wavelets characterization could simulate the response characteristic of primary visual sense preferably.