Research On Design Of Household Intelligent Classification Ashbin Based On Machine Vision

The classification of domestic waste is of great significance for the establishment of a good living environment in the city and the maximization of the use of resources.As the source of household waste generation,the classification and recycling from the family is gradually receiving the attention of relevant government departments and research scholars.Although cities have formulated a policy for waste sorting,due to factors such as lack of classified knowledge and weak classification,the actual classification effect is not satisfactory.With the rapid development of artificial intelligence technology and Internet of Things technology,the intelligentization of garbage classification has become a hot research topic at home and abroad.In view of the current lack of knowledge of household waste classification and the lack of source classification,this paper uses machine vision technology and vgg16 neural network classification algorithm,combined with classification mechanism,to design a family intelligence classification ashbin.The ashbin realize the identification of the types of garbage,and realize automatic classification into the barrel,complete the automatic classification of the garbage instead of the artificial classification to improve the accuracy of the garbage classification.The main research contents of this paper are as follows:1.Through the research and analysis of existing machine vision-based classification devices and recognition classification algorithms,combined with mechanical structure,the overall design of the household intelligent classification trash can is formulated.And based on design requirements,select the hardware that needs to be used throughout the design research process.2.Compare and analyze the existing machine vision system light source settings,image acquisition precautions,etc.,and combine the actual results to determine the image acquisition system camera position,light source position and so on.In addition,combined with the existing classification criteria to determine the classification basis,and collect 13 second-class garbage,a total of 98 kinds of garbage,and collected a total of 1011 pictures through the garbage collection system.3.Data cleaning of the collected images,using VGG16 deep convolutional neural network to establish a garbage recognition model based on machine vision,which can accurately identify the types of garbage,and its recognition accuracy reaches 87.2%;combined with garbage classification Standard,build a model of garbage classification,the final classification accuracy rate reached 88.1%.4.Combine the actual situation of the family to design the overall design of the trash can,determine the appearance style,size parameters,etc.;and design the classification mechanism,use two steering gears to realize the classification of garbage into the barrel.Finally,complete the construction and connection of the overall system,and verify the validity and feasibility of the design through actual cases.