Design and research of intelligent trash can based on convolutional neural network and transfer learning

Abstract

With the improvement of living standards, the types and quantities of garbage produced by residents are increasing, and the garbage disposal methods needs to be further improved. Standardize garbage disposal, garbage classification and garbage recycling will contribute to construction a green city. Garbage classification technologies also needs innovation. The intelligent garbage classification which depends on the Internet and intelligent technologies will overcome the disadvantages of traditional garbage classification. To this end, this paper proposes a smart trash can which based on convolutional neural network and transfer learning technologies. The smart trash can in this paper builds a smart trash monitoring network through NB-IOT, and automatically extract garbage features based on convolutional neural network, then trains the network through transfer learning technologies. The experimental results shows that the smart trash can designed in this paper has fast response speed and high accuracy.