Lightweight safety helmet detection algorithm using improved YOLOv5

• Published in: Journal of real-time image processing Vol. 21; no. 4; p. 125
• Main Authors: Ren, Hongge, Fan, Anni, Zhao, Jian, Song, Hairui, Liang, Xiuman
• Format: Journal Article
• Language: English
• Published: Berlin/Heidelberg Springer Berlin Heidelberg 01.08.2024; Springer Nature B.V
• Subjects: Accuracy; Algorithms; Computation; Computer Graphics; Computer Science; Construction sites; Convolution; Deep learning; Feature extraction; Helmets; Image Processing and Computer Vision; Lightweight; Localization; Machine learning; Modules; Multimedia Information Systems; Object recognition; Parameters; Pattern Recognition; Safety helmets; Sensors; Signal,Image and Speech Processing; Weight reduction; Bottleneck Attention Mechanism; Safety helmet detection; Distribution Shifting Convolution; Lightweight
• ISSN: 1861-8200; 1861-8219
• DOI: 10.1007/s11554-024-01499-5

In response to the challenges faced by existing safety helmet detection algorithms when applied to complex construction site scenarios, such as poor accuracy, large number of parameters, large amount of computation and large model size, this paper proposes a lightweight safety helmet detection algorithm based on YOLOv5, which achieves a balance between lightweight and accuracy. First, the algorithm integrates the Distribution Shifting Convolution (DSConv) layer and the Squeeze-and-Excitation (SE) attention mechanism, effectively replacing the original partial convolution and C3 modules, this integration significantly enhances the capabilities of feature extraction and representation learning. Second, multi-scale feature fusion is performed on the Ghost module using skip connections, replacing certain C3 module, to achieve lightweight and maintain accuracy. Finally, adjustments have been made to the Bottleneck Attention Mechanism (BAM) to suppress irrelevant information and enhance the extraction of features in rich regions. The experimental results show that improved model improves the mean average precision (mAP) by 1.0% compared to the original algorithm, reduces the number of parameters by 22.2%, decreases the computation by 20.9%, and the model size is reduced by 20.1%, which realizes the lightweight of the detection algorithm.