Research on Liquid Level Target Detection of Infusion Bottle Based on Improved YOLOv5
YANG Haochen
GAO Shang
Abstract：In order to better solve the time control of medical staff during patient infusion,advanced depth learning technology is introduced,and based on the depth learning algorithm,target detection and result accuracy optimization are carried out on the liq-uid level position image of infusion bottle.The paper collects the video images of the liquid level height of infusion bottles in differ-ent environments.After preprocessing,the video images are detected based on YOLOv5 algorithm.The target images are trained through GPU.After model training and multiple iterations,and the introduction of SE attention mechanism,a network model with a certain accuracy is obtained.After that,the randomly collected video images are detected in real time.The experimental results show that the improved detection model can better achieve the detection of infusion images.
Keywords：deep learningobject detectionneural network
Publication Date：2025-10-20
Online Publishing Date：2026-01-16(First online date of this platform, not the publication date of the document)
Pages：5( 2787-2790,2830 )