摘要
设计了一种基于神经网络识别的自动跟随行李箱,以树莓派4B作为主控板,在主控板配置TensorflowLite使用环境,并搭载摄像头模块、HC05蓝牙模块和ISD1420语音模块等,通过摄像头获取图像后使用YOLOV3-tiny算法识别人物并使用IOUTracker跟踪算法进行目标跟踪。实验结果表明,该智能行李箱在实际环境能有效的识别使用者并且进行跟踪。
This paper designs an automatic follow-up suitcase based on the neural network recognition.By using the Raspberry Pi 4B as the main control board,configuring the Tensorflow Lite use environment on the main control board,and carrying a camera module,HC05 Bluetooth module and ISD1420 voice module,etc.,people can be identified with the YOLOV3-tiny algorithm after acquiring the image through the camera,and then the IOU Tracker tracking algorithm is used for target tracking.The experimental results show that the smart luggage can effectively identify and track users in the actual environment.
作者
黄明威
林镇炜
肖杰康
杨景康
陈其志
黄辉宇
邓君
HUANG Ming-wei;LIN Zhen-wei;XIAO Jie-kang;YANG Jing-kang;CHEN Qi-zhi;HUANG Hui-yu;DENG Jun(School of Mechanical Engineering,Dongguan University of Technology,Dongguan Guangdong 523000,China)
出处
《机械研究与应用》
2021年第1期115-117,120,共4页
Mechanical Research & Application
关键词
神经网络识别
自动跟随
智能行李箱
neural network recognition
automatic follow
smart luggage