Friday, March 24, 2006
Surveillance Robotics: Using colors to analyse
u0307999 ZHAI NING Among the possible applications which have been foreseen for Service Mobile Robots, surveillance robots (i.e., robots designed to replace human security guards in making rounds) are becoming more and more popular, as witnessed by the many systems commercially available and the growing interest in the research community. In this blog, I just concentrate one area: How is a surveillance robltic can detect unexpected changes in the environment? Well I found out one method from a paper written by Mattia from university of Genova, Italy, which detailed the use of the following mechanism: The robot “looks at” the environment through a TV camera; next, it compares what “it sees” in a specific moment with what “it should see” at that same location. In particular, an approach is proposed for images comparison which requires to find color clusters in the color histograms corresponding to the images to be compared: by analyzing the color clusters in the two images, the system detects similarities or differences between them and consequently deduces if something has changed in the scene. Well that seems like an computer vision problem, and for sure it is. According to another report written by Paolo in the same university, he propose a “ad hoc” algorithms have been implemented for color clusters comparison. The detail is covered in Proceedings 2003 IEEE International Symposium on Computational Intelligence in Robotics and Automation July 1620, 2003, Kobe, Japan. I want to comment on the simply idea: it is only a simplification of the human unconscious way of dection. But it helps a lot for robotic. Can the robotic community do better by consulting more life sciences research result? Maybe it is a good way to start think about.
Posted by Security at 3/24/2006 10:38:00 PM