Personal Protection Equipment detection system for embedded devices based on DNN and Fuzzy Logic

作者：

Highlights：

• Present a real-time Personal Protection Equipment detection system for worker safety.

• Combine Deep learning and Fuzzy to exploit human and machine-learned knowledge.

• Experimental comparison with existing approaches performed on embedded devices.

• Deep Neural Networks + Fuzzy filtering = faster computation and improved performance.

• Require smaller training sets comparing with existing approaches.

摘要

•Present a real-time Personal Protection Equipment detection system for worker safety.•Combine Deep learning and Fuzzy to exploit human and machine-learned knowledge.•Experimental comparison with existing approaches performed on embedded devices.•Deep Neural Networks + Fuzzy filtering = faster computation and improved performance.•Require smaller training sets comparing with existing approaches.

论文关键词：Personal Protection Equipment detection,Deep Neural Networks,Fuzzy Logic,Object detection

论文评审过程：Received 24 January 2021, Revised 20 April 2021, Accepted 12 June 2021, Available online 20 June 2021, Version of Record 1 July 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2021.115447