Search

Safety helmet wearing plays a major role in protecting the safety of workers in industry and construction, so a real-time helmet wearing detection technology is very necessary. This paper proposes an improved YOLOv4 algorithm to achieve real-time and efficient safety helmet wearing detection. The improved YOLOv4 algorithm adopts a lightweight network PP-LCNet as the backbone network and uses deepwise separable convolution to decrease the model parameters. Besides, the coordinate attention mechanism module is embedded in the three output feature layers of the backbone network to enhance the feature information, and an improved feature fusion structure is designed to fuse the target information. In terms of the loss function, we use a new SIoU loss function that fuses directional information to increase detection precision. The experimental findings demonstrate that the improved YOLOv4 algorithm achieves an accuracy of 92.98%, a model size of 41.88 M, and a detection speed of 43.23 pictures/s. Compared with the original YOLOv4, the accuracy increases by 0.52%, the model size decreases by about 83%, and the detection speed increases by 88%. Compared with other existing methods, it performs better in terms of precision and speed. [ABSTRACT FROM AUTHOR]

Edge–edge collaboration is the main collaboration mode for edge computing (EC) in time-sensitive networking (TSN). This distributed collaboration mode, in which the computing work is implemented collaboratively by the edge computing nodes (ECNs) without the involvement of cloud servers, poses challenges to collaboration management. However, most existing methods of ECN collaboration only focus on the scheduling of computing resources rather than management process design and rarely consider the transparency, security, and fair distribution guarantee mechanism of collaboration benefits. To address this issue, we propose a blockchain-based and decentralized management scheme for ECNs autonomous collaboration in TSN. First, we design an edge–edge collaboration management framework and its workflow based on blockchain for EC in TSN (TECChain), which combines blockchain with TSN technology and EC paradigm. To solve the consensus problem in TECChain, the proof of diligence (PoD) and delegated PoD (DPoD) consensus mechanisms are designed. Furthermore, by leveraging the high-precision synchronous clock foundation of TSN, we propose two time-slot-driven consensus algorithms, named sequential decision-making based on DPoD (S-DPoD) and Byzantine fault tolerance based on DPoD (BFT-DPoD) respectively. Theoretical analysis results show that the proposed algorithms perform better in terms of security, transactions per second (TPS), energy saving, etc., compared to other algorithms. Comparison experiments reveal that S-DPoD consumes fewer interactive messages and has higher Byzantine fault tolerance and success rate of consensus than BFT-DPoD, but BFT-DPoD has lower transaction verification latency and larger TPS. [ABSTRACT FROM AUTHOR]