START: A selective thresholding augmented reality toolkit

Augmented Reality (AR) is a new human-computer interaction technology which combines virtual images with physical reality in real time. Most AR systems adopt planar markers providing the positions and the angles information to correctly merge the visual objects. However, planer markers are difficult to detect under uneven illumination conditions or while the markers are partially occluded. In this paper, we proposed a new AR toolkit, named START, to improve the accuracy on marker detection. Two image thresholding methods, dynamical global threshold and adaptive threshold, are provided to cope with the marker detection under different illumination condition. Such a design allows users to select proper thresholding method according to users´ requirements. We also propose a newly designed quadrangle detection and boundary reconstruction method to rebuild the boundaries of occluded markers. A partial pattern matching algorithm is applied to enhance the detection capabilities when the inner pattern of markers are also partially occluded. The result shows that, START can tolerate higher illumination condition, where the measurement of illuminance is twice than the limitation of existing AR toolkits. START can also recognize markers with triple larger occluded area compared to the existing AR toolkits.