Pet Companion System Based on Image Recognition

In order to solve the problem that the pet dog is unattended at home alone and the owner cannot take care of the pet dog under special circumstances, a pet dog companion system based on image recognition is designed. Based on the user's requirement, the system provides various modules such as breed, characteristics, feeding methods and so on. The identification of pet dog breeds is the core module. The deep learning convolution neural network is used to classify pet dogs’ breed accurately. This paper mainly discusses the methods and experiences summarized through deep learning in the process of achieving this module.