Semantic Segmentation Algorithms for Ground AGV and UAV Medical Transport Scenes

The outbreak of the Covid-19 pandemic in recent years and the epidemics of infectious diseases that have occurred around the world over the years, there are problems of lack of medical supplies and difficulties in personnel scheduling. Intelligent medical transportation through modern technology is an effective means to solve this problem. AGV(Automated Guided Vehicle) transportation and UAV(Unmanned Aerial Vehicle) transportation are important ways for intelligent transportation of medical materials. This paper investigates semantic segmentation as a key technology for AGV transport and UAV transport. This paper compares other traditional semantic segmentation networks, and at the same time considers the characteristics of all-weather, all-terrain, and complex transportation of materials in medical transportation, and proposes SSMMTNet(Semantic segmentation of medical material transportation Net). Among them, we propose a Scaling Transformer Block that can extract depth features of point clouds to enrich contextual information. At the same time, the network is validated on the benchmark Semantic3D dataset, obtaining 71.5% mIoU and 90.6% OA.