Medical Waste Classification System Based on OpenCV and SSD-MobileNet for 5G

2021 IEEE Wireless Communications and Networking Conference Workshops (WCNCW) Pub Date : 2021-03-29 DOI:10.1109/WCNCW49093.2021.9420036

Xiaoxiao Bian, Yuchao Chen, Shuyi Wang, Fengyi Cheng, Haotong Cao

{"title":"Medical Waste Classification System Based on OpenCV and SSD-MobileNet for 5G","authors":"Xiaoxiao Bian, Yuchao Chen, Shuyi Wang, Fengyi Cheng, Haotong Cao","doi":"10.1109/WCNCW49093.2021.9420036","DOIUrl":null,"url":null,"abstract":"In the coming 5G era, the medical waste is mostly sorted manually, and medical staff involved in sorting is at great risk of being infected with virus. To solve this problem, a medical waste automatic identification and classification system is designed, which adopt technology of machine vision and deep learning algorithm. The system firstly detects and locates medical waste by cross-platform computer vision library (Open CV), then applies SSD-Mobile Net model to train and classify medical waste. The existing background subtraction algorithm and SSD-MobileNet model structure are improved. Examples of hemostatic forceps, gloves, infusion bags and syringes are tested. Experimental results show that recognition accuracy of medical waste classification system is more than 98.5%, and average recognition time is 52 milliseconds, and the system has strong robustness. The application of the proposed method in actual medical waste classification is of great significance in reducing manual input and reducing the risk of virus infection for medical personnel involved in sorting.","PeriodicalId":262379,"journal":{"name":"2021 IEEE Wireless Communications and Networking Conference Workshops (WCNCW)","volume":"49 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2021-03-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"5","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2021 IEEE Wireless Communications and Networking Conference Workshops (WCNCW)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/WCNCW49093.2021.9420036","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}

引用次数: 5

Abstract

In the coming 5G era, the medical waste is mostly sorted manually, and medical staff involved in sorting is at great risk of being infected with virus. To solve this problem, a medical waste automatic identification and classification system is designed, which adopt technology of machine vision and deep learning algorithm. The system firstly detects and locates medical waste by cross-platform computer vision library (Open CV), then applies SSD-Mobile Net model to train and classify medical waste. The existing background subtraction algorithm and SSD-MobileNet model structure are improved. Examples of hemostatic forceps, gloves, infusion bags and syringes are tested. Experimental results show that recognition accuracy of medical waste classification system is more than 98.5%, and average recognition time is 52 milliseconds, and the system has strong robustness. The application of the proposed method in actual medical waste classification is of great significance in reducing manual input and reducing the risk of virus infection for medical personnel involved in sorting.

查看原文本刊更多论文

基于OpenCV和5G SSD-MobileNet的医疗垃圾分类系统

在即将到来的5G时代，医疗垃圾大多是人工分类，参与分类的医务人员感染病毒的风险很大。为解决这一问题，设计了一种采用机器视觉技术和深度学习算法的医疗垃圾自动识别分类系统。该系统首先通过跨平台计算机视觉库(Open CV)对医疗废物进行检测和定位，然后应用SSD-Mobile Net模型对医疗废物进行训练和分类。对现有的背景减法算法和SSD-MobileNet模型结构进行了改进。测试了止血钳、手套、输液袋和注射器的实例。实验结果表明，该系统的识别准确率达到98.5%以上，平均识别时间为52毫秒，具有较强的鲁棒性。将该方法应用于实际的医疗垃圾分类中，对于参与分类的医务人员减少人工输入，降低病毒感染风险具有重要意义。

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

2021 IEEE Wireless Communications and Networking Conference Workshops (WCNCW)

自引率

0.00%

发文量