期刊名称：Journal of Theoretical and Applied Information Technology
印刷版ISSN：1992-8645
电子版ISSN：1817-3195
出版年度：2014
卷号：70
期号：2
出版社：Journal of Theoretical and Applied
摘要：Laser pointer is a low-cost daily essential item that can provide a prominent spot, deliver the meaning of a pointing gesture and accurately indicate a spot. Robust detection of a laser spot has been made possible with the help of the computer vision techniques. Autonomous laser spot detection system has led to the evolution of many applications such as laser pointer interaction, autonomous robotic helper and simulation based computer game. This paper aims to provide the general architecture and methodology used in a laser spot detection system. Major lower level processing methods, such as thresholding, background subtraction, region of interest extraction and coordinate mapping are fundamental in developing a good laser spot system. Our study found out that most of existing systems are prone to error whenever the lighting condition is varied either gradually or suddenly. The spot is also difficult to detect when it is overlaid on a white background due to the weakness in the detection scheme. Jitter, which is mostly caused by uneven hand movement, is a big challenge in developing an accurate system, especially for laser interaction applications. Finally, we believe that a robust laser spot detection system can be built if the aforementioned challenges are carefully addressed.
关键词：Laser Spot Detection; Coordinate Mapping; ROI Extraction; Thresholding; Background Subtraction
