Skeleton-Based Abnormal Gait Detection
Human gait analysis plays an important role in musculoskeletal disorder diagnosis. Detecting anomalies in human walking, such as shuffling gait, stiff leg or unsteady gait, can be difficult if the prior knowledge of such a gait pattern is not available. We propose an approach for detecting abnormal...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2016-10-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | http://www.mdpi.com/1424-8220/16/11/1792 |
| _version_ | 1818001105719656448 |
|---|---|
| author | Trong-Nguyen Nguyen Huu-Hung Huynh Jean Meunier |
| author_facet | Trong-Nguyen Nguyen Huu-Hung Huynh Jean Meunier |
| author_sort | Trong-Nguyen Nguyen |
| collection | DOAJ |
| description | Human gait analysis plays an important role in musculoskeletal disorder diagnosis. Detecting anomalies in human walking, such as shuffling gait, stiff leg or unsteady gait, can be difficult if the prior knowledge of such a gait pattern is not available. We propose an approach for detecting abnormal human gait based on a normal gait model. Instead of employing the color image, silhouette, or spatio-temporal volume, our model is created based on human joint positions (skeleton) in time series. We decompose each sequence of normal gait images into gait cycles. Each human instant posture is represented by a feature vector which describes relationships between pairs of bone joints located in the lower body. Such vectors are then converted into codewords using a clustering technique. The normal human gait model is created based on multiple sequences of codewords corresponding to different gait cycles. In the detection stage, a gait cycle with normality likelihood below a threshold, which is determined automatically in the training step, is assumed as an anomaly. The experimental results on both marker-based mocap data and Kinect skeleton show that our method is very promising in distinguishing normal and abnormal gaits with an overall accuracy of 90.12%. |
| first_indexed | 2024-04-14T03:30:29Z |
| format | Article |
| id | doaj.art-c8b3c8d958fd4589b4e32465c5347117 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-04-14T03:30:29Z |
| publishDate | 2016-10-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-c8b3c8d958fd4589b4e32465c53471172022-12-22T02:14:59ZengMDPI AGSensors1424-82202016-10-011611179210.3390/s16111792s16111792Skeleton-Based Abnormal Gait DetectionTrong-Nguyen Nguyen0Huu-Hung Huynh1Jean Meunier2DIRO, University of Montreal, Montreal, QC H3T 1J4, CanadaThe University of Danang - University of Science and Technology, Danang 556361, VietnamDIRO, University of Montreal, Montreal, QC H3T 1J4, CanadaHuman gait analysis plays an important role in musculoskeletal disorder diagnosis. Detecting anomalies in human walking, such as shuffling gait, stiff leg or unsteady gait, can be difficult if the prior knowledge of such a gait pattern is not available. We propose an approach for detecting abnormal human gait based on a normal gait model. Instead of employing the color image, silhouette, or spatio-temporal volume, our model is created based on human joint positions (skeleton) in time series. We decompose each sequence of normal gait images into gait cycles. Each human instant posture is represented by a feature vector which describes relationships between pairs of bone joints located in the lower body. Such vectors are then converted into codewords using a clustering technique. The normal human gait model is created based on multiple sequences of codewords corresponding to different gait cycles. In the detection stage, a gait cycle with normality likelihood below a threshold, which is determined automatically in the training step, is assumed as an anomaly. The experimental results on both marker-based mocap data and Kinect skeleton show that our method is very promising in distinguishing normal and abnormal gaits with an overall accuracy of 90.12%.http://www.mdpi.com/1424-8220/16/11/1792human gaitgait analysisgait cyclehidden Markov modelKinect |
| spellingShingle | Trong-Nguyen Nguyen Huu-Hung Huynh Jean Meunier Skeleton-Based Abnormal Gait Detection Sensors human gait gait analysis gait cycle hidden Markov model Kinect |
| title | Skeleton-Based Abnormal Gait Detection |
| title_full | Skeleton-Based Abnormal Gait Detection |
| title_fullStr | Skeleton-Based Abnormal Gait Detection |
| title_full_unstemmed | Skeleton-Based Abnormal Gait Detection |
| title_short | Skeleton-Based Abnormal Gait Detection |
| title_sort | skeleton based abnormal gait detection |
| topic | human gait gait analysis gait cycle hidden Markov model Kinect |
| url | http://www.mdpi.com/1424-8220/16/11/1792 |
| work_keys_str_mv | AT trongnguyennguyen skeletonbasedabnormalgaitdetection AT huuhunghuynh skeletonbasedabnormalgaitdetection AT jeanmeunier skeletonbasedabnormalgaitdetection |