Goose Surface Temperature Monitoring System Based on Deep Learning Using Visible and Infrared Thermal Image Integration

Owing to increased biosecurity and industrial demands, the poultry houses in Taiwan are generally nonopen closed types, with automatic environmental control sensor equipment gradually being installed such houses. Environmental sensors health monitoring systems necessary improve feeding efficiency safety. In this work, we developed a goose surface temperature system based on deep learning using visible image integrated infrared thermal image. This could detect geese obtain individual automatically. consisted of an embedded trained detection model, camera, camera. The Mask R-convolutional neural network algorithm was employed train model by collected images. camera captured images house, which be identified model. temperatures were obtained through integration land pool areas commercial house monitor achieved precision 97.1% recall 95.1%. addition, area observed lower than that area. would used as management index managers.