Research On Intelligent Classification Of Urban Waste Cans Based On Deep Learning

With the rapid development of science and technology,people’s living standards have been greatly improved,and garbage in urban life is also growing rapidly.A large amount of garbage is transported to the city for landfill or incineration.Only part of the garbage is disposed harmlessly.The speed of garbage disposal is slow and the level of intelligent garbage classification is low.How to realize the rapid classification of garbage has become an unavoidable problem for the country and even the whole world.In this paper,an intelligent classification design of urban garbage bin based on in-depth learning is proposed.Specific methods are as follows: building an intelligent garbage bin monitoring network through NB-IOT Internet of Things technology,building an intelligent garbage classification model and algorithm using convolutional neural network.The simulation results show that the classification algorithm has fast response and high accuracy.This paper mainly analyses the garbage classification at home and abroad,and discusses the research status of intelligent garbage bin;constructs the remote monitoring network of garbage bin based on NB-IOT,designs the overall structure of garbage bin and the circuit of hardware control system;in software,preprocesses the garbage image,completes the feature extraction of garbage image,fuses the features of texture,shape and HSV color features are used as input samples of BP neural network to complete the training of BP neural network.On the other hand,the three-layer network structure after Alexnet convolution neural network model is modified,and the garbage RGB image is used as input samples of Alexnet convolution neural network for migration learning,thus the garbage intelligent classification model is constructed.The simulation experiment of intelligent garbage classification based on MATLAB is carried out.The simulation results of intelligent garbage classification based on convolution neural network are compared with traditional intelligent classification methods such as BP neural network.The classification and discrimination experiments of 90 test samples are carried out.The results show that the accuracy of the migrated convolution neural network for garbage image classification reaches 100%,and that of the traditional BP neural network.Comparedwith image classification(about 70% accuracy),the training speed of the migrated convolutional neural network is faster and the recognition rate is higher.The results show that the migration learning algorithm of convolution neural network has potential application value in intelligent classification of garbage images.Figure [57] table [8] reference [50].