PROCESSING METHOD FOR SMART DECISION-BASED TARGET DETECTION MODEL, AND RELATED DEVICE

摘要

A processing method for a smart decision-based target detection model, an apparatus, a computer device, and a storage medium, the method comprising: on the basis of an acquired local dataset, training an initial multi-target detection model to obtain an intermediate multi-target detection model (S202); on the basis of generated additional random numbers and model parameters of the intermediate multi-target detection model, calculating composite model parameters (S203); sending the composite model parameters to a central server to indicate to the central server to generate global model parameters on the basis of the composite model parameters (S204); receiving the global model parameters to update the intermediate multi-target detection model (S205); and causing the updated intermediate multi-target detection model to serve as an initial multi-target detection model for a next round of training and performing iterative training until model convergence to obtain a multi-target detection model (S206). In addition, the present invention further relates to blockchain technology, and the local data set can be stored in a blockchain. The accuracy of target detection is thus improved.