Garbage object recognition and classification based on Mask Scoring RCNN
Shuijing Li, Ming Yan, Jie Xu
Abstract
In order to protect the ecological environment on which human beings depend for existence and make the society sustainable development, it is more and more necessary to classify the garbage. However, people are not familiar with the classification method, so it is difficult for us to accurately understand the classification of each kind of garbage. In order to guide people to classify garbage correctly, this paper proposes a garbage classification system based on Mask Scoring RCNN. Make a new data set based on Beijing municipal domestic garbage classification criteria, and use Mask Scoring RCNN for training. It is hoped that this system can help people classify garbage accurately and reduce the waste of time caused by garbage classification. The test results show that the garbage classification system in this paper has a high accuracy, and the segmentation recognition effect is better in the case of complex picture, which can efficiently and conveniently complete the garbage classification guidance task.