Device and method for automatically classifying garbage

Abstract

The invention relates to a device and method for automatically classifying garbage. The device comprises a machine body, wherein a microprocessor is mounted in the machine body, and a trained improved ResNet-50 garbage classification convolutional neural network model is stored in the microprocessor; the machine body is provided with a throwing opening for throwing garbage, the throwing opening is provided with a camera for image acquisition, and the camera is connected with the input end of the microprocessor; four different types of garbage cans are arranged on the machine body, are respectively a recyclable garbage can, a kitchen garbage can, a harmful garbage can and an other garbage can, and are used for classifying different types of garbage; a conveyor belt is arranged on the machine body and located under the feeding opening, and a driving motor is arranged below the conveyor belt and used for running of the conveyor belt. The conveyor belt is connected with the output end of the microprocessor and used for conveying the garbage output by the microprocessor into the corresponding can. According to the invention, automatic identification of garbage can be realized automatically, accurately and efficiently; and the garbage can be automatically put into the corresponding garbage can.

Description

technical field [0001] The invention relates to the technical field of garbage automatic classification, in particular to a device and method for automatic garbage classification. Background technique [0002] The ResNet-50 network uses multiple very small convolution kernels to replace a single large-sized convolution kernel, which can not only reduce the parameters of the ResNet-50 network, but also increase the number of nonlinear activation functions of the ResNet-50 network and reduce the number of ResNet-50 network parameters. 50 network calculations. According to the task characteristics of garbage classification and identification, it can be seen that garbage is divided into four categories, and each category of garbage contains many different types of garbage, such as cans, mineral water bottles, etc., which itself contains many different types of characteristics. There is a certain difficulty in extracting detailed features. [0003] In reality, garbage sorting t...

AI Technical Summary

Problems solved by technology

[0003] In reality, garbage sorting technology uses garbage cans of different garbage categories to be placed side by side. The limitation is that people need to judge the garbage category by themselves and manually sort garbage; the functions of some relatively intelligent garbage cans are limited to: realizing automatic opening, Automatic packaging, etc., did not realize the function of automatic identification and classification of garbage

Images