Road Garbage Classification Using ResNet50

Nomika Sree Kolla, Mythili Anumula, S. Sujana, M. Ratnababu
{"title":"Road Garbage Classification Using ResNet50","authors":"Nomika Sree Kolla, Mythili Anumula, S. Sujana, M. Ratnababu","doi":"10.1109/IConSCEPT57958.2023.10170073","DOIUrl":null,"url":null,"abstract":"Nowadays managing the road garbage is essential. The waste disposal leads to pollution, climate change, water contamination etc. The major issue which is unresolved is dealing with the large amount of waste that is dumped in the environment rather than segregating properly. To overcome this problem, a deep learning algorithm is used to segregate the garbage which is beneficial for diminishing landfills, recycling etc. We use One MaxPool layer, one average pool layer, and 48 convolutional layers make up the 50-layer convolutional neural network known as ResNet50. The model that is used to categorise the items has already been trained. In the process of implementation certain stages are involved such as preprocessing, DataAugmentation, training, Finetuning and evaluation of the modal etc. This work aims to keep the environment safe and also helps the municipal corporations to collect garbage effectively in remote areas. The garbage dataset consists of 2,527 images of cardboard, plastic, paper, metal, glass and trash. We achieved an accuracy of 81%. Finally, Precision, Recall, f1 scores and Confusion matrix are calculated with respect to their classes.","PeriodicalId":240167,"journal":{"name":"2023 International Conference on Signal Processing, Computation, Electronics, Power and Telecommunication (IConSCEPT)","volume":"19 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2023-05-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2023 International Conference on Signal Processing, Computation, Electronics, Power and Telecommunication (IConSCEPT)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/IConSCEPT57958.2023.10170073","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

Nowadays managing the road garbage is essential. The waste disposal leads to pollution, climate change, water contamination etc. The major issue which is unresolved is dealing with the large amount of waste that is dumped in the environment rather than segregating properly. To overcome this problem, a deep learning algorithm is used to segregate the garbage which is beneficial for diminishing landfills, recycling etc. We use One MaxPool layer, one average pool layer, and 48 convolutional layers make up the 50-layer convolutional neural network known as ResNet50. The model that is used to categorise the items has already been trained. In the process of implementation certain stages are involved such as preprocessing, DataAugmentation, training, Finetuning and evaluation of the modal etc. This work aims to keep the environment safe and also helps the municipal corporations to collect garbage effectively in remote areas. The garbage dataset consists of 2,527 images of cardboard, plastic, paper, metal, glass and trash. We achieved an accuracy of 81%. Finally, Precision, Recall, f1 scores and Confusion matrix are calculated with respect to their classes.
使用ResNet50进行道路垃圾分类
如今,管理道路垃圾是必不可少的。垃圾处理导致污染、气候变化、水污染等。尚未解决的主要问题是如何处理倾倒在环境中的大量废物,而不是进行适当的分类。为了克服这一问题,采用深度学习算法对垃圾进行分类,有利于减少垃圾填埋、回收利用等。我们使用一个MaxPool层,一个平均池层和48个卷积层组成50层卷积神经网络,称为ResNet50。用于对物品进行分类的模型已经经过了训练。在实施过程中,涉及到预处理、数据增强、训练、微调和模态评估等阶段。这项工作旨在保护环境安全,并帮助市政公司有效地收集偏远地区的垃圾。垃圾数据集包括2527张纸板、塑料、纸张、金属、玻璃和垃圾的图像。我们达到了81%的准确率。最后,计算精度、召回率、f1分数和混淆矩阵。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信