Research on helmet wearing detection method based on deep learning.

The vigorous development of the construction industry has also brought unprecedented safety risks. The wearing of safety helmets at the construction site can effectively reduce casualties. As a result, this paper suggests employing a deep learning-based approach for the real-time detection of safety helmet usage among construction workers. Based on the selected YOLOv5s network through experiments, this paper analyzes its training results. Considering its poor detection effect on small objects and occluded objects. Therefore, multiple attention mechanisms are used to improve the YOLOv5s network, the feature pyramid network is improved into a BiFPN bidirectional feature pyramid network, and the post-processing method NMS is improved into Soft-NMS. Based on the above-improved method, the loss function is improved to enhance the convergence speed of the model and improve the detection speed. We propose a network model called BiFEL-YOLOv5s, which combines the BiFPN network and Focal-EIoU Loss to improve YOLOv5s. The average precision of the model is increased by 0.9% the recall rate is increased by 2.8%, and the detection speed of the model does not decrease too much. It is better suited for real-time safety helmet object detection, addressing the requirements of helmet detection across various work scenarios.

A GC×GC-MS method based on solid-state modulator for non-targeted metabolomics: Comparison with traditional GC-MS method.

The formidable challenge posed by the presence of extremely high amounts of compounds and large differences in concentrations in plasma significantly complicates non-targeted metabolomics analyses. In this study, a comprehensive two-dimensional gas chromatography-quadrupole mass spectrometry (GC×GC-qMS) method with a solid-state modulator (SSM) for non-targeted metabolomics in beagle plasma was first established based on a GC-MS method, and the qualitative and quantitative performance of the two platforms were compared. Identification of detected compounds was accomplished utilizing NIST database match scores, retention indices (RIs) and standards. Semi-quantification involved the calculation of peak area ratios to internal standards. Metabolite identification sheets were generated for plasma samples on both analytical platforms, featuring 22 representative metabolites chosen for validating qualitative accuracy, and for conducting comparisons of linearity, accuracy, precision, and sensitivity. The outcomes revealed a threefold increase in the number of identifiable metabolites on the GC×GC-MS platform, with lower limits of quantitation (LLOQs) reduced to 0.5-0.05 times those achieved on the GC-MS platform. Accuracy in quantification for both GC×GC-MS and GC-MS fell within the range of 85-115%, and the vast majority of intra- and inter-day precisions were within the range of 20%. These findings underscore that relative to the conventional GC-MS method, the GC×GC-MS method developed in this study, combined with SSM, exhibits enhanced qualitative capabilities, heightened sensitivity, and comparable accuracy and precision, rendering it more suitable for non-targeted metabolomics analyses.