摘要
为了提高垃圾智能分拣的效率和准确度,该文设计了基于机器视觉的垃圾自动分拣装置。装置首先利用金属传感器识别金属类垃圾,然后采用YOLOv3算法对其余垃圾进行目标检测,并使用ResNet101卷积神经网络模型进行分类,训练得到的目标检测模型及图像分类模型部署在Intel i38145 UE平台。装置采用两个步进电机分别控制托盘和隔板,提高垃圾的投放效率。垃圾分类测试结果表明,该装置实现了垃圾的自动分类和投放,对4类垃圾的平均识别准确率为98.3%,平均投放时间为3.1 s,具有一定的推广应用价值。
In order to improve the efficiency and accuracy of intelligent garbage sorting,this paper designs an automatic garbage sorting device based on machine vision.The device first uses metal sensors to identify metal garbage,then uses YOLOv3 algorithm to detect the remaining garbage,and uses ResNet101 convolutional neural network model to classify.The trained target detection model and image classification model are deployed on the Intel i38145 UE platform.The device uses two stepper motors to control the tray and the partition respectively to improve the efficiency of garbage discharge.The test results of garbage classification show that the device realizes the automatic classification and delivery of garbage.The average recognition accuracy of 4 types of garbage is 98.3%,and the average delivery time is 3.1 s,which has certain application value.
作者
付丽
梁华林
冯若愚
秦瑞
李语桐
FU Li;LIANG Hualin;FENG Ruoyu;QIN Rui;LI Yutong(Key Lab of Optoelectronic Technology&Systems of Ministry of Education,Chongqing University,Chongqing 400044,China;School of Big Data&Software Engineering,Chongqing University,Chongqing 401331,China;CQU-UC Joint CO-OP Institute,Chongqing University,Chongqing 400044,China)
出处
《实验技术与管理》
CAS
北大核心
2023年第1期123-127,159,共6页
Experimental Technology and Management
基金
教育部产学合作协同育人项目(202101355071)
教育部高等学校光电教指分委教育教学研究项目(2020SYL20)
重庆市教育委员会教育教学改革研究项目(CQGJ19B01)。
