DocumentCode
996291
Title
A 2-D motion detection model for low-cost embedded reconfigurable I/O devices
Author
Dollas, Apostolos ; Sotiropoulos, Stamatios ; Papademetriou, Kyprianos
Author_Institution
Electron. & Comput. Eng. Dept., Tech. Univ. of Crete, Chania, Greece
Volume
52
Issue
8
fYear
2005
Firstpage
1443
Lastpage
1449
Abstract
A low-cost reconfigurable embedded apparatus for two-dimensional (2-D) motion detection has been developed. This paper briefly outlines the embedded reconfigurable system architecture, and presents in-depth the 2-D motion detection model, which is directly mapped to reconfigurable hardware. Emphasis is placed on the hardware ability to adapt to individual needs of kinetically challenged persons by altering detection thresholds and delays, thus resulting into an efficient low-cost reconfigurable hardware implementation of the model. This paper also presents how the model detects complex motions through a vocabulary of simple motions, and how the system is trained to individual users´ needs. Experimental results and integrated applications of the model for text processing are also presented.
Keywords
biomechanics; biomedical equipment; input-output programs; medical signal detection; text analysis; word processing; 2-D motion detection model; complex motion detection; kinetically challenged persons; low-cost embedded reconfigurable I/O devices; text processing; Control systems; Embedded system; Field programmable gate arrays; Hardware; Microcontrollers; Microprocessors; Motion detection; Programmable logic arrays; Virtual reality; Vocabulary; 2-D motion detection; Input/output (I/O) device; kinetically challenged; reconfigurable embedded system; Electronics, Medical; Equipment Design; Equipment Failure Analysis; Humans; Microcomputers; Movement; Movement Disorders; Therapy, Computer-Assisted; Transducers; User-Computer Interface;
fLanguage
English
Journal_Title
Biomedical Engineering, IEEE Transactions on
Publisher
ieee
ISSN
0018-9294
Type
jour
DOI
10.1109/TBME.2005.851485
Filename
1463333
Link To Document