An Improved Magnetic Tracking Approach Based on ResNet-LM Fusion Algorithm

Guo, Pengfei; Dai, Houde; Yang, Qianhui; Huang, Qiaoyuan; Yao, Hanchen

doi:10.1109/tim.2022.3216597

Cited by 4 publications

(2 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…(2) The proposed features constructed in the BSN are too simple to capture enough context in time. (3) The BSN is multi-stage without a unified framework.…”

Section: Preliminariesmentioning

confidence: 99%

See 1 more Smart Citation

A falls recognition framework based on faster R-CNN and temporal action sequences that can be deployed on home service robots

Mei,

Liu,

2024

Meas. Sci. Technol.

View full text Add to dashboard Cite

With the popularity of home service robots (e.g., floor sweepers), robots should be considered to have more elderly care features. Compared to fixed home monitors with a limited field of view, falls detection with service robots is an ideal solution to keep the elderly and disabled people within sight. However, the user's actions such as lying on the bed to sleep or slumping on the sofa to rest, cause the traditional falls detection system to generate false alarms, which disrupts the user's family life. In this article, the improved Faster R-CNN network was proposed by adding temporal action sequences and calculating fall acceleration, which shows the low misjudgment rate on the service robot platform. Firstly, motion images were captured to obtain the target's motion area description and action timing at the input stage. Then, the Faster R-CNN algorithm was implemented to further check the suspected falls based on the falling acceleration of the detected actions during the training phase. Finally, the action mistaken for falling was eliminated by the proposed temporal action sequences module. Network training and robotic platform test results show the proposed method can recognize falls (or false falls) to avoid false alarms. On the service robot platform, experimental results show the FAR is 8.19 and the processing time is 0.79s.

show abstract

“…(2) The proposed features constructed in the BSN are too simple to capture enough context in time. (3) The BSN is multi-stage without a unified framework.…”

Section: Preliminariesmentioning

confidence: 99%

“…Failure to detect falls and provide medical assistance may have fatal consequences in terms of the older adult's death. If someone lies on the floor more than an hour after falling, his/her chance of dying within six months is 50% [3]. Prediction or early detection of a fallen person and subsequent actions to call for help are vital.…”

Section: Introductionmentioning

confidence: 99%