Logistics park safety helmet wearing detection and segmentation method and system and medium

The invention discloses a logistics park safety helmet wearing detection and segmentation method and system and a medium, and relates to the technical field of computer vision. On the basis of a DeepLabV3 + network model structure, a GridAttene-Net semantic segmentation network model is constructed, and the model comprises an expansion convolution residual network feature extraction module, a convolution attention mechanism fusion module and an image feature gridding module. A ResNet-101 expansion residual feature extraction network is adopted as a backbone extraction network, so that the calculation amount is reduced, and the model convergence speed of the network is increased. A convolution attention mechanism fusion module is introduced in a coding stage, the characterization capability of a feature region is effectively enhanced, and the precision of image semantic segmentation is further improved. And meanwhile, the image feature gridding module can obtain small target feature information of a local area