RGB camera-based fallen person detection system embedded on a mobile platform

作者：

Highlights：

• Low-cost assistive robot for elderly care centers can detect falls in real-time.

• A publicly dataset to fairly compare fallen person vision-based algorithms.

• A deep learning model differentiates between fallen persons and resting postures.

• A facial recognition algorithm is adapted for lying postures in different angles.

摘要

•Low-cost assistive robot for elderly care centers can detect falls in real-time.•A publicly dataset to fairly compare fallen person vision-based algorithms.•A deep learning model differentiates between fallen persons and resting postures.•A facial recognition algorithm is adapted for lying postures in different angles.

论文关键词：Assistive robot,Fallen person detection,Face recognition,Object detector,Convolutional neural network,Support vector machine

论文评审过程：Received 12 August 2020, Revised 3 December 2021, Accepted 20 February 2022, Available online 28 February 2022, Version of Record 8 March 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.116715