Low-altitude drone detection method based on environmental antiinterference

Conference: ICMLCA 2021 - 2nd International Conference on Machine Learning and Computer Application
12/17/2021 - 12/19/2021 at Shenyang, China

Proceedings: ICMLCA 2021

Pages: 5Language: englishTyp: PDF

Personal VDE Members are entitled to a 10% discount on this title

Authors:
Bai, Fan; Sun, Haoyang; Cao, Zhaorui (Equipment Engineering College, Shenyang Ligong University, Shenyang, China)
Zhang, Hui (Mechanical Engineering College, Shenyang Ligong University, Shenyang, China)

Abstract:
In order to achieve real-time detection and rapid warning of low, small and slow UAV targets in severe weather, in this paper we combine multiple cascaded autoencoders, with the attention mechanism, to remove rain streaks in the input images by using multi-stage information exchange and feature fusion. We use ResNet as feature extraction network, to quickly obtain the category and position information of the UAV by predicting the target center point, so as to realize the real-time early warning for small drone targets. The experimental results show that the average recognition accuracy of the proposed algorithm can reach 82% under the heavy rain weather, and the calculation speed can reach 24 frames per second, which can effectively prevent the penetration of non-cooperative UAVs in the airspace.