{"title":"A weight pruned convolutional neural network for garbage image classification","authors":"Yuchen Wang, Guangtai Zhao, Xiaoguang Li","doi":"10.1109/CISP-BMEI56279.2022.9980320","DOIUrl":null,"url":null,"abstract":"Developing automatic garbage sorting technologies will be helpful to promote the construction of ecological civilization in China. A weighted convolutional neural network for the categorization of garbage images. First, a garbage image classification algorithm is implemented using a convolutional neural network. Then, the architecture of the network is simplified by weight pruning to get rid of the dependency of GPU servers. In addition, a garbage image database has been established to train and test the proposed algorithm. The experimental results show that the accuracy of proposed method can reach 90.88% on the test set, and the running time of processing each image takes 78ms.","PeriodicalId":198522,"journal":{"name":"2022 15th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2022-11-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2022 15th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/CISP-BMEI56279.2022.9980320","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
Developing automatic garbage sorting technologies will be helpful to promote the construction of ecological civilization in China. A weighted convolutional neural network for the categorization of garbage images. First, a garbage image classification algorithm is implemented using a convolutional neural network. Then, the architecture of the network is simplified by weight pruning to get rid of the dependency of GPU servers. In addition, a garbage image database has been established to train and test the proposed algorithm. The experimental results show that the accuracy of proposed method can reach 90.88% on the test set, and the running time of processing each image takes 78ms.