现代计算机2024，Vol.30Issue(24) ：77-80.DOI:10.3969/j.issn.1007-1423.2024.24.014

基于深度学习的多模态无人机视觉识别手势指令实现飞行控制的方法研究

Research on flight control method of multimodal UAV visual recognition gesture command based on deep learning

蒋方园
现代计算机2024，Vol.30Issue(24) ：77-80.DOI:10.3969/j.issn.1007-1423.2024.24.014
✕
  • 万方数据

基于深度学习的多模态无人机视觉识别手势指令实现飞行控制的方法研究

Research on flight control method of multimodal UAV visual recognition gesture command based on deep learning

蒋方园1
扫码查看

作者信息

  • 1. 武汉铁路职业技术学院计算机与信息工程学院,武汉 430074
  • 折叠

摘要

随着无人机技术的迅速发展,基于手势的控制方式在无人机自主飞行领域受到广泛关注.研究采用多模态融合技术设计一种高效的无人机手势控制算法.实验结果显示,研究算法的手势识别准确率高达0.98,召回率达0.96,响应时间最低为102.1 ms,飞行精度最高为92.54%.该方法在手势识别和飞行控制的准确性与稳定性方面具有显著优势.

Abstract

With the rapid development of UAV technology,gesture based control has been widely concerned in the field of UAV autonomous flight.An efficient gesture control algorithm for UAV is designed by using multimodal fusion technology.The ex-perimental results show that the accuracy of gesture recognition is as high as 0.98,the recall rate is 0.96,the minimum response time is 102.1 ms,and the maximum flight accuracy is 92.54%.This method has significant advantages in the accuracy and stability of gesture recognition and flight control.

关键词

无人机/手势识别/多模态融合/深度学习/实时响应

Key words

UAV/gesture recognition/multimodal fusion/deep learning/real time response

引用本文复制引用

出版年

2024
现代计算机
中大控股

现代计算机

影响因子：0.292
ISSN：1007-1423
段落导航相关论文