An investigation into the closure process in human visual perception for line patterns

摘要

Closure property can provide a key to the image understanding in human visual perception. It is a fundamental intuitive property of pattern grouping as advocated by the Gestalt school of psychology. The closure possesses a property to organize the world by filling in the gaps in the stimulation so that we perceive a whole object and not disjointed parts (organization principle). To investigate the organization principle in human visual perception for line patterns, closure process and the function of visual receptive field are considered in this paper. Two corresponding methods from the viewpoint of computer science are presented and applied to the removal of the closure noise appearing in the binary DPC (degraded paint character) image. One is called DE (disk expansion) method based on the boundary information of stimulation to simulate the closure process. The other is called VF (visual receptive field) method based on the lateral spread of excitation in neurophysiology to provide the effective VF for closure process. The combination of DE and VF methods is also developed. A series of detailed experiments are given to illustrate the proposed algorithms and to compare the performance of the proposed algorithms with that of related existing methods (morphology method and relaxation method). Experiments show that the proposed algorithms can obtain the better smoothing results and can reasonably simulate the closure process in human visual perception for line patterns.