Computer Vision For The Blind

Conclusions
Conclusion In this project it was shown how indoor blind navigation is conceivable through a system based on tracking feature points and relating these to a prior generated map of the building. This method was based on processing inputs from a video camera and inertial sensor data to track the required feature points. Maps of commercial and residential buildings will be required for this method to eventually be successfully implemented. This is a significant task but was successfully completed for outdoor GPS. The eventual outcome, assuming that a majority of buildings are mapped successfully would enable blind persons to freely navigate through commercial buildings of their choosing without the requirement of further assistance other than that provided by this system.

Conclusions

Conclusion

In this project it was shown how indoor blind navigation is conceivable through a system based on tracking feature points and relating these to a prior generated map of the building. This method was based on processing inputs from a video camera and inertial sensor data to track the required feature points. Maps of commercial and residential buildings will be required for this method to eventually be successfully implemented. This is a significant task but was successfully completed for outdoor GPS. The eventual outcome, assuming that a majority of buildings are mapped successfully would enable blind persons to freely navigate through commercial buildings of their choosing without the requirement of further assistance other than that provided by this system.

Copyright © Christopher Nielsen, 2009. All rights reserved. Contact: collectorchris@shaw.ca