A Method for Eliminating Pig Face Recognition Errors Caused by Too Short Pig Growth Cycle

In the process of modern large-scale pig breeding, it is necessary to distinguish identity each and real-time detect its health status, weight change, dietary other parameters. Traditional methods waste a lot resources, while high quality pork cannot be effectively guaranteed. This project based on convolutional neural networks design develop face recognition system. system uses an overhead camera suspended above house monitor for 24 hours identify track pig. Due rapid growth cycle pig, facial image information changes rapidly, which has significant impact model. The acquisition module designed correct monitoring tracking data. installed in place every day collect upload collected server. By comparing data individual pigs at different developmental stages with server, identification determined, corrected timely manner. At same time, screen displayed screen, behavioral parameters are recorded facilitate management breeding farm.