8 February 2017 Behavioral similarity measurement based on image processing for robots that use imitative learning
Author Affiliations +
Proceedings Volume 10225, Eighth International Conference on Graphic and Image Processing (ICGIP 2016); 102251Y (2017) https://doi.org/10.1117/12.2267035
Event: Eighth International Conference on Graphic and Image Processing, 2016, Tokyo, Japan
Abstract
In the field of the artificial societies, particularly those are based on memetics, imitative behavior is essential for the development of cultural evolution. Applying this concept for robotics, through imitative learning, a robot can acquire behavioral patterns from another robot. Assuming that the learning process must have an instructor and, at least, an apprentice, the fact to obtain a quantitative measurement for their behavioral similarity, would be potentially useful, especially in artificial social systems focused on cultural evolution. In this paper the motor behavior of both kinds of robots, for two simple tasks, is represented by 2D binary images, which are processed in order to measure their behavioral similarity. The results shown here were obtained comparing some similarity measurement methods for binary images.
© (2017) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Dante G. Sterpin B., Fernando Martinez S., Edwar Jacinto G., "Behavioral similarity measurement based on image processing for robots that use imitative learning ", Proc. SPIE 10225, Eighth International Conference on Graphic and Image Processing (ICGIP 2016), 102251Y (8 February 2017); doi: 10.1117/12.2267035; https://doi.org/10.1117/12.2267035
PROCEEDINGS
6 PAGES


SHARE
Back to Top