Pilot Hand Detection Based on Semi-Supervised Learning

In this paper, we propose a pilot hand detection method that combines the semi-supervised DenSe Learning (DSL) with a coordinate attention (CA) module, which we call CA-DSL. In our suggested approach, the CA module is added to Resnet, the backbone network of DSL. The CA module enhances the DSL’s capacity for learning and boosts the effectiveness of pilot hand detection. In addition, we establish a dataset for pilot hand detection that we call Pilot Hand (PH). On the PH dataset, we evaluate the suggested hand detection algorithm. and compared it with the state-of-the-art models. Experimental results show that the detection accuracy of CA-DSL outperforms the state-of-the-art models by 2%-7%.