Low, Tobias and Manzanera, Antoine (2010) Ground-plane classiﬁcation for robot navigation: combining multiple cues toward a visual-based learning system. In: ICARCV 2010: 11th International Conference on Control, Automation, Robotics and Vision , 7-10 Dec 2010, Singapore.
This paper describes a vision-based ground-plane classiﬁcation system for autonomous indoor mobile-robot that takes advantage of the synergy in combining together multiple visual-cues. A priori knowledge of the environment is important in many biological systems, in parallel with their reactive systems. As such, a learning model approach is taken here for the classiﬁcation of the ground/object space, initialised through a new Distributed-Fusion (D-Fusion) method that captures colour and textural data using Superpixels. A Markov Random Field (MRF) network is then used to classify, regularise, employ a priori constraints, and merge additional ground/object information provided by other visual cues (such as motion) to improve classiﬁcation images. The developed system can classify indoor test-set ground-plane surfaces with an average true-positive to false-positive rate of 90.92% to 7.78% respectively on test-set data. The system has been designed in mind to fuse a variety of different visual-cues. Consequently it can be customised to ﬁt different situations and/or sensory architectures accordingly.
|Item Type:||Conference or Workshop Item (Commonwealth Reporting Category E) (Paper)|
|Additional Information:||Author version not held.|
|Uncontrolled Keywords:||image classiﬁcation; image disparity; ground plane; obstacle avoidance; visual navigation; mobile robots|
|Depositing User:||Dr Tobias Low|
|Date Deposited:||16 Mar 2011 05:25|
|Last Modified:||02 Oct 2013 03:20|
Actions (login required)
|Archive Repository Staff Only|