Mask wearing detection method fusing 3D attention mechanism and cavity convolution

• Main Authors: TIAN CHENGJUN, LIU HAOBO, LIU ZHE, WANG YUYU, ZHANG JINTONG, YAN YU
• Format: Patent
• Language: Chinese; English
• Published: 23.06.2023
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; PHYSICS

The invention discloses a mask wearing detection method fusing a 3D attention mechanism and cavity convolution, relates to the technical field of machine vision target detection, solves the problem that a detection method which is high in training speed, high in detection precision and high in detection speed needs to be provided in the prior art, and comprises the steps: marking whether mask wearing is standard or not for a first data set; an improved YOLOv5 deep learning network model is established; pre-training the improved YOLOv5 deep learning network model by using a second data set to obtain an optimal pre-training model weight; using a transfer learning mode, using the optimal pre-training model weight to initialize parameters of the improved YOLOv5 deep learning network model, using the first data set to train the model, and obtaining a mask wearing detection model. According to the method, the two indexes of detection precision and detection speed are balanced, the detection precision of the model i