改进Inception-Resnet-V2网络的无人机航向识别

doi:10.3778/j.issn.1002-8331.2106-0542

摘要/Abstract

摘要： 为解决无人机在复杂环境下电力巡检的避障难题，研究并改进了基于Inception-Resnet-V2网络的一种无人机航向识别方法。引入深度可分离卷积，将卷积操作分解为深度卷积和逐点卷积两个过程，压缩了计算量。改进后的网络结构保证高精度的识别，同时节约了计算成本。改进后的网络模型在标准数据集上达到了92.5%的准确率。在实际电力巡检实验中，改进的网络模型针对于基杆塔的航向预测精度达到95.63%。实验结果表明，搭载改进后Inception-Resnet-V2网络模型的无人机可以在复杂环境下成功识别大型基杆塔并进行精确地航向识别与预测。

关键词: 图像识别, 卷积神经网络, 可分离卷积, 航向识别, Inception-Resnet-V2网络

Abstract: In order to solve the obstacle avoidance issue of UAV power patrol inspection in complex environment, a UAV heading recognition method based on Inception-Resnet-V2 network is studied and improved. The depth separable convolution is introduced, and the convolutional operation is decomposed into two processes：depth convolution and point-by-point convolution, which compresses the amount of calculation. The improved network structure ensures high-precision recognition and saves the calculation cost. The improved network model achieves an accuracy of 92.5% on the standard data set. In the actual power inspection experiment, the heading prediction accuracy of the improved network model for the base tower is 95.63%. Experimental results show that UAV equipped with the improved Inception-Resnet-V2 network model can successfully identify large base towers and accurately identify and predict the heading in complex environments.

Key words: image recognition, convolutional neural network, separable convolution, course identification, Inception-Resnet-V2 network

成怡, 田文斌, 郑腾龙. 改进Inception-Resnet-V2网络的无人机航向识别[J]. 计算机工程与应用, 2022, 58(24): 307-312.

CHENG Yi, TIAN Wenbin, ZHENG Tenglong. Unmanned Aerial Vehicles Heading Recognition Based on Improved Inception-Resnet-V2 Network[J]. Computer Engineering and Applications, 2022, 58(24): 307-312.

参考文献

[1] 刘壮，杜勇，陈怡.±500kV直流输电线路直线塔无人机巡检安全距离仿真与试验[J].高电压技术，2019，45（2）：426-432.
LIU Z，DU Y，CHEN Y.Simulation and experiment on the safety distance of typical ±500 kV DC transmission lines and towers for UAV inspection[J].High Voltage Engineering，2019，45（2）：426-432.
[2] 王盛誉，毛鹏飞，谢晓君.智能识别技术在无人机电力巡检中的应用[J].集成电路应用，2021，38（3）：98-99.
WANG S Y，MAO P F，XIE X J.Application of intelligent recognition technology in UAV power inspection[J].Applications of IC，2021，38（3）：98-99.
[3] 邓能辉，吴创嘉，杨朝霖.中厚板表面缺陷检测关键技术的研究与应用[J].冶金自动化，2021，45（5）：86-92.
DENG N H，WU C J，YANG C L.Research and application of key technology for surface defects detection of steel plates[J].Metallurgical Industry Automation，2021，45（5）：86-92.
[4] 袁培森，申成吉，徐焕良.基于迁移学习和双线性Inception-ResNet-v2的细粒度菌菇识别方法：201910547744.8[P].
2019-09-10.
YUAN P S，SHEN C J，XU H L.A fine-grained mushroom identification method based on migration learning and bilinear Inception-ResNet-v2：201910547744.8[P].2019-09-10.
[5] WANG Y J，GUAN Q，LAO I W.Using deep convolutional neural networks for multi-classification of thyroid tumor by histopathology：a large-scale pilot study[J].Annals of Translational Medicine，2019，7（18）：468.
[6] ZHU Y C，ZHANG C K.Generating research of image caption based on improved NIC algorithm[J].Journal of Computational Methods in Sciences and Engineering，2018，19（2）：353-366.
[7] SZEGEDY C，LIU W，JIA Y Q，et al.Going deeper with convolutions[C]//2015 IEEE Conference on Computer Vision and Pattern Recognition.Boston：IEEE Press，2015：1-9.
[8] LIN M，CHEN Q，YAN S C.Network in network[EB/OL].（2013-12-16）.https：//arxiv.org/abs/1312.4400.
[9] GRAHA B，MAATEN L.Submanifold sparse convolutional networks[EB/OL].（2017-06-05）.https：//arxiv.org/abs/1706.
01307.
[10] SZEGEDY C，IOFFE S，VANHOUCKE V.Inception-v4，Inception-ResNet and the impact of residual connections on learning[J].arXiv：1602.07261，2016.
[11] HE K M，ZHANG X Y，REN S Q，et al.Deep residual learning for image recognition[C]//29th IEEE Conference on Computer Vision and Pattern Recognition.Las Vegas：IEEE Computer Society，2016：770-778.
[12] HOWARD A G，ZHU M，CHEN B.MobileNets：efficient convolutional neural networks for mobile vision applications[J].arXiv：1704.04861，2017.
[13] KAISER L，GOMEZ A N，CHOLLET F.Depthwise separable convolutions for neural machine translation[J].arXiv：1706.03059，2017.
[14] KRIZHEVSKY A，SUTSKEVER I，HINTON G.ImageNet classification with deep convolutional neural networks[J].Communications of the ACM，2017，60（6）：84-90.
[15] SIONYAN K，ZISSERAN A.Very deep convolutional networks for large-scale image recognition[J].arXiv：1409.
01556，2014.
[16] CHOLLET F.Xception：deep learning with depthwise separable convolutions[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition.Honolulu：IEEE Press，2017：1800-1807.