Author:Department of Bio-Industrial Mechatronics Engineering, NTU
Pig husbandry accounts for 46.1% of the domestic animal husbandry output value, and the output value is as high as NT$75.558 billion. Among all the development stages, piglets are relatively vulnerable and need more attention.
This project proposes to automatically monitor piglet activities using embedded systems and deep learning. The embedded systems comprise Raspberry Pi, cameras, and microphones. The videos recorded by the embedded systems are used to train deep convolutional neural network models. Therefore, the trained model was able to notice workers in pig farm immediately when abnormal behavior happened. Moreover, to do statistics on the observed within one month for the workers.