Garbage classification method and system based on transfer learning and model fusion and medium

Abstract

The invention discloses a garbage classification method and system based on transfer learning and model fusion and a storage medium. In the method, the method includes classifying the junk image datathrough a classifier, and outputting a corresponding classification result; moreover, the generation process of the classifier adopted by the invention comprises the step of constructing two differentclassification networks into two strong classifiers by using an Adaboost algorithm, so that the generalization error rate is low, over-fitting is avoided, and the high accuracy of industrial requirements can be achieved. And the two strong classifiers are fused in a decision-making layer fusion mode, so that the feature diversity is fully utilized, and the classification accuracy is further improved.

Description

technical field [0001] The present invention relates to the fields of image analysis and recognition and machine learning, in particular to a garbage classification method, system and medium based on transfer learning and model fusion, and belongs to the image target detection technology based on deep learning. Background technique [0002] As the concept of low-carbon and environmentally friendly life becomes more and more popular, the problem of garbage classification has attracted more and more people's attention. According to the plan of the Ministry of Housing and Urban-Rural Development, by the end of 2020, 46 key cities, including Beijing, Chengdu, and Chongqing, should basically realize the waste sorting and processing system, and other prefecture-level cities should realize full coverage of domestic waste sorting in public institutions. However, the definition of garbage classification is vague. For example, melon seed shells are officially identified as wet garbage...

AI Technical Summary

Problems solved by technology

However, such methods are highly complex, usually only applicable to certain specific scenarios, have poor generalization ability, and are easily disturbed by environmental factors in the actual application process, resulting in the disadvantage of low recognition accuracy in complex environments.

Images