Intelligent garbage classification system and method based on machine vision

Abstract

The invention discloses an intelligent garbage classification system and method based on machine vision. By applying MobileNet deep learning and machine vision technology, Raspberry pi, STM32 are adopted as a main controller, a conveying device is designed and implemented, full-automatic loading of garbage is achieved, and manual loading is not needed. According to testing results, key performanceof the system is greatly improved compared with classifying garbage cans on the market, and the system successfully achieves the effects that (1) the system can adapt to different temperature and lighting environments, judgment is accurate; (2) the recognition precision is increased to 95% or above; and (3) excellent processing capacity is achieved, detection of up to 128 points can be completedat a high speed within 7 seconds, and the recognition speed is high.

Description

technical field [0001] The invention relates to an intelligent garbage classification system and method based on machine vision. Background technique [0002] With the development of economy and society, the amount of domestic waste generated in my country has increased rapidly, and the resulting environmental problems have become increasingly prominent, which has become a restrictive factor for the development of new urbanization. Garbage classification is an important link and key field in the construction of ecological civilization. my country is forcing the establishment of a domestic waste classification management system, and various provinces and cities have responded one after another. Changsha City will also officially implement waste classification in October this year. Before 2019, 8 cities in my country have carried out pilot projects of waste sorting and collection, but the popularization effect is not satisfactory. The reasons why waste classification cannot ...

AI Technical Summary

Problems solved by technology

Before 2019, 8 cities in my country have carried out pilot projects of waste sorting and collection, but the popularization effect is not satisfactory

However, the few existing system solutions at home and abroad are not very practical and have relatively single functions. The common problems include: (1) the recognition accuracy is not high, resulting in classification errors; (2) the recognition speed is too slow; (3) the temperature is affected , lighting and other environmental impacts are obvious, and the performance is unstable; (4) lack of detailed data analysis, it is difficult to efficiently connect with the recycling link

Among them, the low recognition accuracy and slow recognition speed are the key pain points that make this type of trash can difficult to use

Images