Abstract | ||
---|---|---|
Identification of the evacuees with walking difficulty will definitely lead to quick rescue and thus improve the efficiency of evacuation in times of disasters or calamities. We are developing a new method using singular value decomposition for extracting features from the time-series data which is measured with various sensors such as an accelerometer, a motion capture system and a force sensor. In this paper, we apply this method to assess walking difficulty based on three dimensional acceleration data during walking. In order to verify the usefulness of the method, three levels of walking disability in the lower limbs are simulated by constraining the knee joint and ankle joint of the right leg. The accelerations of the middle of shanks and the back of the waist are measured and analyzed after normalization. Features related to walking difficulty are acquired from the time-series acceleration data using singular value decomposition. The results showed that the first singular values inferred from the acceleration data of the right and left shanks significantly related to the increase of the constraint to the joints. The first singular values of the shanks were suggested to be reliable criteria to evaluate walking difficulty. We propose a triangular tool to provide intuitive information extracted from the first singular values to assist the evaluation of the walking difficulty. |
Year | DOI | Venue |
---|---|---|
2012 | 10.1109/SCIS-ISIS.2012.6505376 | SCIS&ISIS |
Keywords | Field | DocType |
accelerometers,data visualisation,disasters,emergency services,feature extraction,force sensors,gait analysis,singular value decomposition,time series,accelerometer,ankle joint,calamities,dimensional acceleration data,evacuee walking difficulty evaluation,evacuee walking difficulty visualizaton,force sensor,knee joint,leg,lower limbs,motion capture system,shanks,singular values,time-series acceleration data,time-series data,walking disability | Motion capture,Normalization (statistics),Singular value,WALKING DIFFICULTY,Computer science,Artificial intelligence,Ankle,Singular value decomposition,Computer vision,Simulation,Accelerometer,Acceleration,Machine learning | Conference |
ISSN | ISBN | Citations |
2377-6870 | 978-1-4673-2742-8 | 0 |
PageRank | References | Authors |
0.34 | 4 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Isao Hayashi | 1 | 276 | 85.75 |
Yinlai Jiang | 2 | 22 | 7.01 |
Shuoyu Wang | 3 | 89 | 27.69 |