Fusion along time-dimension to investigate aspects of visual pattern completion

Image fusion along time dimension is investigated to explain aspects of multisensors visual patterns and the related representation tasks. We address image fusion problem by describing a layered complex channels neural model. In processing visual data conventionally complex channels utilize both the linear and nonlinear operations, and consist of two stages filtering. The first filtering is sensitive to higher spatial frequencies than the second, and they suppose to explain a number of imaging task complexities that can not be explained by simple channels. The work reported here focuses on practical applications and behavior modeling to examine new perspectives for complex processes at the sensory vision that begins with antagonistic decomposition of input. This is then followed by fusion of emergent time-sensitive data to promote a nonlinear dynamic association. Our reported result suggests that time fusion of data in multistaged complex channels of neural processes exhibit the true adaptive behavior that adjust itself to a changing environment.