使用Yolov5和KCF进行人群计数

Q1 Engineering Periodicals of Engineering and Natural Sciences Pub Date : 2023-03-24 DOI:10.21533/pen.v11i2.3490

M. A. J. Maktoof, Ibraheem Nadher Ibraheem, Israa Tahseen Ali Al_attar

{"title":"使用Yolov5和KCF进行人群计数","authors":"M. A. J. Maktoof, Ibraheem Nadher Ibraheem, Israa Tahseen Ali Al_attar","doi":"10.21533/pen.v11i2.3490","DOIUrl":null,"url":null,"abstract":"Crowd detection has various applications nowadays. However, detecting humans in crowded circumstances is difficult because the features of different objects conflict, making cross-state detection impossible. Detectors in the overlapping zone may therefore overreact. In this paper, real-time people counting is proposed using a proposed model of the YOLOv5 (You Only Look Once) algorithm and KCF (kernel correlation filter) algorithm. The YOLOv5 algorithm was used because it is considered one of the most accurate algorithms for detecting people in real time. Despite the high accuracy of the YOLOv5 algorithm in detecting the people in the image, video or real-time camera capturing, it needs an increase in speed. For this reason, the YOLOv5 algorithm was combined with the KCF tracking algorithm. Where the YOLOv5 algorithm identifies people to be tracked by the KCF. The YOLOv5 algorithm was trained on a database of people, and the system's accuracy reached 98%. The speed of the proposed system was increased after adding the KCF.","PeriodicalId":37519,"journal":{"name":"Periodicals of Engineering and Natural Sciences","volume":" ","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2023-03-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Crowd counting using Yolov5 and KCF\",\"authors\":\"M. A. J. Maktoof, Ibraheem Nadher Ibraheem, Israa Tahseen Ali Al_attar\",\"doi\":\"10.21533/pen.v11i2.3490\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Crowd detection has various applications nowadays. However, detecting humans in crowded circumstances is difficult because the features of different objects conflict, making cross-state detection impossible. Detectors in the overlapping zone may therefore overreact. In this paper, real-time people counting is proposed using a proposed model of the YOLOv5 (You Only Look Once) algorithm and KCF (kernel correlation filter) algorithm. The YOLOv5 algorithm was used because it is considered one of the most accurate algorithms for detecting people in real time. Despite the high accuracy of the YOLOv5 algorithm in detecting the people in the image, video or real-time camera capturing, it needs an increase in speed. For this reason, the YOLOv5 algorithm was combined with the KCF tracking algorithm. Where the YOLOv5 algorithm identifies people to be tracked by the KCF. The YOLOv5 algorithm was trained on a database of people, and the system's accuracy reached 98%. The speed of the proposed system was increased after adding the KCF.\",\"PeriodicalId\":37519,\"journal\":{\"name\":\"Periodicals of Engineering and Natural Sciences\",\"volume\":\" \",\"pages\":\"\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2023-03-24\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Periodicals of Engineering and Natural Sciences\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.21533/pen.v11i2.3490\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"Engineering\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Periodicals of Engineering and Natural Sciences","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.21533/pen.v11i2.3490","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"Engineering","Score":null,"Total":0}

引用次数: 0

摘要

人群检测现在有各种各样的应用。然而，在拥挤环境中检测人类是困难的，因为不同物体的特征相互冲突，使得跨状态检测成为不可能。因此，重叠区域的探测器可能反应过度。本文提出了一种基于YOLOv5 (You Only Look Once)算法和KCF (kernel correlation filter)算法的实时人口统计模型。之所以使用YOLOv5算法，是因为它被认为是实时检测人员最准确的算法之一。尽管YOLOv5算法在检测图像、视频或实时摄像机捕捉中的人物方面具有很高的准确性，但它需要提高速度。为此，我们将YOLOv5算法与KCF跟踪算法相结合。其中YOLOv5算法确定KCF要跟踪的人。YOLOv5算法在人的数据库上进行了训练，系统的准确率达到98%。在加入KCF后，系统的速度得到了提高。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

查看原文

微信好友朋友圈 QQ好友复制链接

本刊更多论文

Crowd counting using Yolov5 and KCF

Crowd detection has various applications nowadays. However, detecting humans in crowded circumstances is difficult because the features of different objects conflict, making cross-state detection impossible. Detectors in the overlapping zone may therefore overreact. In this paper, real-time people counting is proposed using a proposed model of the YOLOv5 (You Only Look Once) algorithm and KCF (kernel correlation filter) algorithm. The YOLOv5 algorithm was used because it is considered one of the most accurate algorithms for detecting people in real time. Despite the high accuracy of the YOLOv5 algorithm in detecting the people in the image, video or real-time camera capturing, it needs an increase in speed. For this reason, the YOLOv5 algorithm was combined with the KCF tracking algorithm. Where the YOLOv5 algorithm identifies people to be tracked by the KCF. The YOLOv5 algorithm was trained on a database of people, and the system's accuracy reached 98%. The speed of the proposed system was increased after adding the KCF.

求助全文

通过发布文献求助，成功后即可免费获取论文全文。去求助

来源期刊

Periodicals of Engineering and Natural Sciences Engineering-Architecture

CiteScore

1.90

自引率

0.00%

发文量

140

审稿时长

7 weeks

期刊介绍： *Industrial Engineering: 1 . Ergonomics 2 . Manufacturing 3 . TQM/quality engineering, reliability/maintenance engineering 4 . Production Planning 5 . Facility location, layout, design, materials handling 6 . Education, case studies 7 . Inventory, logistics, transportation, supply chain management 8 . Management 9 . Project/operations management, scheduling 10 . Information systems for production and management 11 . Innovation, knowledge management, organizational learning *Mechanical Engineering: 1 . Energy 2 . Machine Design 3 . Engineering Materials 4 . Manufacturing 5 . Mechatronics & Robotics 6 . Transportation 7 . Fluid Mechanics 8 . Optical Engineering 9 . Nanotechnology 10 . Maintenance & Safety *Computer Science: 1 . Computational Intelligence 2 . Computer Graphics 3 . Data Mining 4 . Human-Centered Computing 5 . Internet and Web Computing 6 . Mobile and Cloud computing 7 . Software Engineering 8 . Online Social Networks *Electrical and electronics engineering 1 . Sensor, automation and instrumentation technology 2 . Telecommunications 3 . Power systems 4 . Electronics 5 . Nanotechnology *Architecture: 1 . Advanced digital applications in architecture practice and computation within Generative processes of design 2 . Computer science, biology and ecology connected with structural engineering 3 . Technology and sustainability in architecture *Bioengineering: 1 . Medical Sciences 2 . Biological and Biomedical Sciences 3 . Agriculture and Life Sciences 4 . Biology and neuroscience 5 . Biological Sciences (Botany, Forestry, Cell Biology, Marine Biology, Zoology) [...]