| Title | Behavior Analysis based on Local Object Tracking and its Bed-exit Application |
| Publication Type | Conference Paper |
| Year of Publication | 2021 |
| Authors | Shih, Chi-Huang, Lin, Cheng-Jian, Wei, Ta-Sen, Liu, Peng-Ta, Shih, Ching-Yu |
| Conference Name | 2021 IEEE 4th International Conference on Knowledge Innovation and Invention (ICKII) |
| Keywords | bed exit, behavior analysis, composability, Error analysis, infrared sensors, Metrics, object oriented security, object tracking, privacy, Prototypes, pubcrawl, Resiliency, surveillance, Technological innovation |
| Abstract | Human behavior analysis is the process that consists of activity monitoring and behavior recognition and has become the core component of intelligent applications such as security surveillance and fall detection. Generally, the techniques involved in behavior recognition include sensor and vision-based processing. During the process, the activity information is typically required to ensure a good recognition performance. On the other hand, the privacy issue attracts much attention and requires a limited range of activity monitoring accordingly. We study behavior analysis for such privacy-oriented applications. A local object tracking (LOT) technique based on an infrared sensor array is developed in a limited monitoring range and is further realized to a practical bed-exit system in the clinical test environment. The experimental results show a correct recognition rate of 99% for 6 bedside activities. In addition, 89% of participants in a satisfaction survey agree on its effectiveness. |
| DOI | 10.1109/ICKII51822.2021.9574741 |
| Citation Key | shih_behavior_2021 |
Behavior Analysis based on Local Object Tracking and its Bed-exit Application