Segmentation model of dorsal hand vein based on improved U-Net
Xiaonan Gao, Guangyuan Zhang, Kang Wang · Proceedings of the 2021 5th International Conference on Electronic Information Technology and Computer Engineering · 2021
Since the outbreak of COVID-19, more than 230, 000 medical staff worldwide have been infected with the new crown virus. The non-artificially contacted dorsal hand vein automatic injection method has been welcomed by many medical workers because of its good isolation. The key to realizing non-contact automatic injection of dorsal hand veins is to realize the detection and segmentation of dorsal hand veins and the decision-making of the needle point position. In this article, aiming at the problem of dorsal hand vein detection, a semantic segmentation model based on U-Net's improved guidance and attention mechanism is proposed. Based on the self-built dorsal hand vein database, the test has achieved good dorsal hand vein detection results.