Garbage classification system and method based on transfer learning

Abstract

The invention provides a garbage classification method based on transfer learning. The garbage classification method comprises the following steps: constructing a convolutional neural network; obtaining a garbage image training data set; performing parameter optimization on the convolutional neural network through a simulated annealing algorithm; training the parameter-optimized convolutional neural network through the garbage image training data set; and identifying and classifying garbage through the trained convolutional neural network. According to the method, the problems of large workload and low efficiency due to the fact that garbage classification in China mainly adopts manual sorting at present can be solved, the training time can be shortened, and the accuracy can be improved. Correspondingly, the invention further provides a garbage classification system based on transfer learning.

Description

technical field [0001] The present invention relates to the technical field of image recognition, in particular to a garbage classification system and method based on migration learning. Background technique [0002] According to "How Much Garbage 2.0" released by the World Bank in 2018, China produced about 220 million tons of urban waste in 2016, of which about 60% could be reused. It can be seen that garbage recycling has huge ecological and economic benefits, and if we want to make full use of garbage resources or reduce the generation of garbage, the premise is to classify garbage. [0003] At present, the garbage sorting in my country is mainly based on manual sorting, which has the problems of heavy workload and low efficiency. Therefore, it is necessary to develop an intelligent garbage sorting method. Contents of the invention [0004] Based on this, in order to solve the problems of heavy workload and low efficiency due to manual sorting of garbage sorting in my...

AI Technical Summary

Problems solved by technology

[0004] Based on this, in order to solve the problems of heavy workload and low efficiency due to manual sorting of garbage sorting in my country, the present invention provides a garbage sorting system and method based on transfer learning. The specific technical solutions are as follows:

Images