基于深度神经网络的低空弱小无人机目标检测研究 -- 西北工业大学学报,2018,36(2):258-263
论文:2018,Vol:36,Issue(2):258-263
引用本文:
王靖宇, 王霰禹, 张科, 蔡宜伦, 刘越. 基于深度神经网络的低空弱小无人机目标检测研究[J]. 西北工业大学学报
Wang Jingyu, Wang Xianyu, Zhang Ke, Cai Yilun, Liu Yue. Small UAV Target Detection Model Based on Deep Neural Network[J]. Northwestern polytechnical university

基于深度神经网络的低空弱小无人机目标检测研究
王靖宇1,2, 王霰禹1,2, 张科1,2, 蔡宜伦1,2, 刘越1,2
1. 航天飞行动力学技术重点实验室, 陕西 西安 710072;
2. 西北工业大学 航天学院, 陕西 西安 710072
摘要:
针对低空无人机目标视觉特征较弱,传统识别模型在目标尺度较小时易受干扰导致识别精度下降等问题,提出了一种基于多隐含层深度神经网络的弱小无人机目标检测模型。根据低空监视图像输入特性和弱小无人机目标视觉表征特点,设计了包含多个隐含层的多通道深度神经网络模型结构,并通过建立多尺度、多角度、多背景条件下的无人机目标图像数据库,完成了对深度网络模型参数的训练及优化。仿真结果表明,所设计的深度模型对低空无人机目标具有较好的变尺度检测能力和抗干扰效果,体现出良好的鲁棒性和潜在工程应用前景。
关键词:    低空无人机    目标识别    深度神经网络    多隐含层   
Small UAV Target Detection Model Based on Deep Neural Network
Wang Jingyu1,2, Wang Xianyu1,2, Zhang Ke1,2, Cai Yilun1,2, Liu Yue1,2
1. National Key Laboratory of Aerospace Flight Dynamics, Xi'an 710072, China;
2. School of Astronautics, Northwestern Polytechnical University, Xi'an 710072, China
Abstract:
Unmanned aerial vehicle (UAV) has relatively small size and weak visual characteristics. The recognition accuracy of traditional object detection methods can decrease sharply when complex background and distraction objects exist. In this paper, we proposed a novel deep neural network (DNN) model for small UAV target recognition task. Based on the visual characteristics of surveillance image and UAV target, a multi-channel DNN is designed. Training and optimization of the DNN are completed with self-constructed UAV image database. Simulation results show that the proposed DNN model can achieve good results in recognizing the variable-scale UAV target and have compatible performance in distinguishing the interference and that the proposed model is robust and has a great potential prospect for engineering application.
Key words:    unmanned aerial vehicle(UAV)    object recognition    deep neural network(DNN)    multi-hidden layer    neural networks    optimization   
收稿日期: 2017-06-12     修回日期:
DOI:
基金项目: 国家自然科学基金(61174204,61101191,61502391)、航天支撑基金(N2015KC0121)、航天飞行动力学技术重点实验室开放基金、陕西省自然科学基础研究计划(2017JM6043)与中央高校基本科研业务费资助
通讯作者:     Email:
作者简介: 王靖宇(1985-),西北工业大学讲师,主要从事智能图像处理及探测导引技术研究。
相关功能
PDF(1594KB) Free
打印本文
把本文推荐给朋友
作者相关文章
王靖宇  在本刊中的所有文章
王霰禹  在本刊中的所有文章
张科  在本刊中的所有文章
蔡宜伦  在本刊中的所有文章
刘越  在本刊中的所有文章

参考文献:
[1] 蔡亚梅,姜宇航,赵霜. 国外反无人机系统发展动态与趋势分析[J]. 航天电子对抗,2017,33(2):59-64 Cai Yamei, Jiang Yuhang, Zhao Shuang. Development Status and Trend Analysis of Counter UAV Systems[J]. Aerospace Electronic Warfare, 2017,33(2):59-64(in Chinese)
[2] 李国军. 民用无人机恐怖袭击风险评估与防御策略研究[J]. 中国人民公安大学学报:社会科学版,2017,33(3):9-14 Li Guojun. Study on Risk Assessment and Defensive Strategy of Terrorist Attacks on Civil UAV[J]. Journal of People's Public Security University of China, Social Sciences Edition, 2017, 33(3):9-14(in Chinese)
[3] Wang A, Lu J, Cai J, et al, Large-Margin Multi-Modal Deep Learning for RGB-D Object Recognition[J]. IEEE Trans on Multimedia, 2015, 17(11):1887-1898
[4] Zhu Yingying, Zhang Chengquan, Zhou Duoyou, et al. Traffic Sign Detection and Recognition Using Fully Convolutional Network Guided Proposals[J]. Neurocomputing, 2016, 214(suppl C):758-766
[5] Takeki, Akito, Tu Tuan Trinh, Ryota Yoshihashi, et al. Combining Deep Features for Object Detection at Various Scales:Finding Small Birds in Landscape Images[J]. IPSJ Trans on Computer Vision and Applications 20168(1):5
[6] 蒋兆军,成孝刚,彭雅琴,等. 基于深度学习的无人机识别算法研究[J]. 电子技术应用,2017,43(7):84-87 Jiang Zhaojun, Cheng Xiaogang, Peng Yaqin, et al. A Novel UAV Recognition Algorithm Based on Deep Learning Approach[J]. Application of Electronic Technique, 2017,43(7):84-87(in Chinese)
[7] Rozantsev, Artem, Vincent Lepetit, Pascal Fua. Detecting Flying Objects Using a Single Moving Camera.[J]. IEEE Trans on Pattern Analysis and Machine Intelligence 2017, 39(5):879-92
[8] Rawat W, Wang Z. Deep Convolutional Neural Networks for Image Classification:A Comprehensive Review[J]. Neural Computation, 2017, 29(9):2352-2449