基于深度学习的多分类垃圾识别系统设计与实现OA
Design and implementation of multi-classification garbage recognition system based on deep learning
针对当前垃圾检测存在多分类垃圾检测效率较低且分类不准确的情况,设计并实现了一种垃圾检测系统,该系统基于深度学习YOLOv11n技术并结合PyQt5,系统运用YOLOv11n模型,借助大量标注数据开展训练,达成了对玻璃垃圾、金属垃圾、塑料垃圾以及其他垃圾的高精度识别.此系统支持图片、视频以及摄像头输入,可实时进行目标检测,实验结果显示,系统在不同场景下都可以稳定运行,检测准确率最高可达94%,切实提升了垃圾检测与分类效率,为智能垃圾分类给予了高效的技术支撑.
In view of the low efficiency and inaccurate classification of multi-classification garbage detection in the current garbage detection,a garbage detection system is designed and implemented.The system is based on deep learning YOLOv11n technology com-bined with PyQt5.The system uses the YOLOv11n model and uses a large number of labeled data to carry out training.Achieved high-precision identification of glass garbage,metal garbage,plastic garbage and other garbage.This system supports picture,video and cam-era input,and can perform target detection in real time.The experimental results show that the system can run stably in different scenari-os,and the maximum detection accuracy rate can reach 94%,which effectively improves the efficiency of garbage detection and classifica-tion.Garbage classification provides efficient technical support.
刘权;闫金萌;喻恒;周晓钰;朱英豪
平顶山学院信息工程学院,河南 平顶山 467000平顶山学院信息工程学院,河南 平顶山 467000平顶山学院信息工程学院,河南 平顶山 467000平顶山学院信息工程学院,河南 平顶山 467000平顶山学院信息工程学院,河南 平顶山 467000
信息技术与安全科学
图像识别深度学习YOLOv11nPyQt5垃圾分类实时检测
Image recognitionDeep learningYOLOv11nPyQt5Garbage classificationReal-time detection
《通信与信息技术》 2026 (1)
62-66,5
平顶山学院2025年人工智能教学改革研究与实践专项项目(项目编号:RGZN202503)平顶山学院大学生创新创业训练项目(项目编号:109192024072)
评论