Computer vision on a colour blindness plate

摘要

An approach is presented to computer vision on a colour blindness plate (CBP), which is a satisfactory way of testing colour vision. To effectively recognize the meaningful dotted pattern (e.g. a figure) from a complex dotted image (e.g. CBP), based on human visual perception, a passive and an active process are principally designed for the proposed approach. The passive process is used to perform image segmentation for a CBP. The active process is used to perform the generalization and matching for pattern recognition. In this paper, the algorithms for each process are presented and exemplified with a series of illustrations. This approach is applied to demonstrating normal colour vision, as well as red-green and total colour blindness.