删除或更新信息,请邮件至freekaoyan#163.com(#换成@)

一种基于深度学习的高速无人艇视觉检测实时算法

本站小编 Free考研考试/2021-12-21

本文二维码信息
二维码(扫一下试试看!)
一种基于深度学习的高速无人艇视觉检测实时算法
A Real-Time Algorithm for Visual Detection of High-Speed Unmanned Surface Vehicle Based on Deep Learning
投稿时间:2020-07-30
DOI:10.15918/j.tbit1001-0645.2018.317
中文关键词:高速无人艇深度学习视觉检测实时
English Keywords:high-speed unmanned surface vehicledeep learningvisual detectionreal-time
基金项目:
作者单位
周治国北京理工大学 信息与电子学院, 北京 100081
刘开元北京理工大学 信息与电子学院, 北京 100081
郑翼鹏北京理工大学 信息与电子学院, 北京 100081
屈崇北京理工大学 信息与电子学院, 北京 100081
中国船舶重工集团公司第七一一研究所, 上海 201108
王黎明中国人民解放军海军工程大学, 湖北, 武汉 430032
摘要点击次数:510
全文下载次数:362
中文摘要:
针对高速无人艇自主航行时对视觉检测的实时性以及由于水面场景变化和波浪反射等干扰的鲁棒性需求,提出一种基于深度学习的高速无人艇视觉检测实时算法.采用基于MobileNet的神经网络快速提取全图特征,设计SSD结构的检测网络融合各层特征图的结果以完成快速多尺度检测,并在嵌入式GPU NVIDIA Jetson TX2硬件平台上将算法实现并验证.结果表明,该算法能够实时检测多类水上特定目标,具有鲁棒性强、多尺度的特点,单帧视频的检测时间可以控制在50 ms以内.
English Summary:
In order to solve the problem of high real-time, and robustness to disturbances such as scene changes and wave reflections during visual detection in unmanned surface vehicle(USV) autonomous navigation, a real-time algorithm for visual detection of high-speed USV based on deep learning was proposed. First, a neural network MobileNet was arranged to quickly extract the full-image features. Then, a detection network based on SSD was used to fuse feature maps of each layer and achieve fast and multi-scale detection. Finally, the algorithm was implemented and validated on a hardware platform embedded GPU NVIDIA Jetson TX2. The results show that the proposed algorithm can quickly detect multiple types of specific obstacle on the water with strong robustness and multi-scale detection ability, and the detection speed of single-frame video within 50 ms.
查看全文查看/发表评论下载PDF阅读器
相关话题/信息 电子 北京 北京理工大学 视觉