Real Time Object Detection Based on Deep Neural Network

Tarek Teama, Hongbin Ma*, Ali Maher, Mohamed A. Kassab

*此作品的通讯作者

科研成果: 书/报告/会议事项章节会议稿件同行评审

1 引用 (Scopus)

摘要

In this research we focus on using deep learning for the training of real time detection of defected Nails and Nuts on a high speed production line using You Only Look Once (YOLO) algorithm for real time object detection and trying to increase the precision of detection and decrease the problems facing real time object detection models like Object occlusion, different orientation for objects, lighting conditions, undetermined moving objects and noise. A series of experiments have been done to achieve high prediction accuracy, the experimental results made on our costumed pascal visual object classes (VOC) dataset demonstrated that the mean Average Precision (mAP) could reach 85%. The proposed model showed very good prediction accuracy on the test dataset.

源语言英语
主期刊名Intelligent Robotics and Applications - 12th International Conference, ICIRA 2019, Proceedings
编辑Haibin Yu, Jinguo Liu, Lianqing Liu, Yuwang Liu, Zhaojie Ju, Dalin Zhou
出版商Springer Verlag
493-504
页数12
ISBN(印刷版)9783030275372
DOI
出版状态已出版 - 2019
活动12th International Conference on Intelligent Robotics and Applications, ICIRA 2019 - Shenyang, 中国
期限: 8 8月 201911 8月 2019

出版系列

姓名Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
11743 LNAI
ISSN(印刷版)0302-9743
ISSN(电子版)1611-3349

会议

会议12th International Conference on Intelligent Robotics and Applications, ICIRA 2019
国家/地区中国
Shenyang
时期8/08/1911/08/19

指纹

探究 'Real Time Object Detection Based on Deep Neural Network' 的科研主题。它们共同构成独一无二的指纹。

引用此