Protective Gloves Reflective Clothing Detection Algorithm Based on Improved Yolov8n

Electric power work is a high-risk field with frequent accidents, so the safety of power workers is of paramount importance. In order to more effectively manage and monitor power operations, reduce accident risks, and facilitate equipment deployment, this study proposes a reflective clothing and protective gloves detection algorithm based on the improved YOLOv8n model. The core ideas of this algorithm include the following three points: (1) Replace the Darknet part of the YOLOv8n model with the ShuffleNetV2 structure to significantly reduce the amount of parameters while maintaining detection accuracy. (2) Introduce the RFB module to enhance the receptive field in the feature extraction process while integrating information of different depths. (3) Add the Triplet Attention mechanism to the feature extraction part to promote the establishment of interdependence between features, thereby improving the perceptual ability of the model. Experimental results show that compared with the original YOLOv8n model, the improved model has significantly improved the detection accuracy of protective gloves and reflective clothing, and its average precision has increased by 2.6%, confirming the practicability of the algorithm. and efficiency. At the same time, the parameter amount and computational complexity (GFLOPs) of the improved model were reduced by 10.1% and 14.6% respectively, effectively achieving fast and accurate detection of protective gloves and reflective clothing. Such improvements provide strong support for the safety and efficiency of power operations.