Abstract
We propose an evolvable hardware (EHW) architecture for pattern recognition. Compared to an earlier offline EHW pattern recognition approach, the proposed architecture is advantageous in its suitability for online adaptation while maintaining a very high recognition speed. With its support for virtual run-time reconfiguration, the architecture is suitable for implementation in an on-chip evolution system. Function level modules and data buses are employed in the architecture in order to reduce the search space. Furthermore, incremental evolution is applied, which shortens evolution time and allows for the evolution of a larger system. The use of incremental evolution also has the advantage of reducing the size of the evolution hardware in an on-chip evolution system. Variations in architecture parameters are explored, and it is found that the size of the system can be reduced at the cost of longer evolution time or lower recognition accuracy. The architecture is applied to a face image recognition task, for which a recognition accuracy of 96.25% is demonstrated. This result is better than the previously proposed offline EHW architectures.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
