[1]黄伟,曹宇剑,徐国明.基于Faster R-CNN模型的低空平台偏振高光谱目标检测[J].红外技术,2019,41(7):600-606.[doi:10.11846/j.issn.1001_8891.201907002]
 HUANG Wei,CAO Yujian,XU Guoming.Polarized Hyperspectral Object Detection with Faster R-CNN for Low-Altitude Platforms[J].Infrared Technology,2019,41(7):600-606.[doi:10.11846/j.issn.1001_8891.201907002]
点击复制

基于Faster R-CNN模型的低空平台偏振高光谱目标检测
分享到:

《红外技术》[ISSN:1001-8891/CN:CN 53-1053/TN]

卷:
41卷
期数:
2019年第7期
页码:
600-606
栏目:
出版日期:
2019-07-20

文章信息/Info

Title:
Polarized Hyperspectral Object Detection with Faster R-CNN for Low-Altitude Platforms

文章编号:
1001-8891(2019)07-0600-07
作者:
黄伟1曹宇剑2徐国明23
1. 中国电子科技集团公司第二十七研究所;
2. 中国人民解放军陆军炮兵防空兵学院;
3. 安徽新华学院信息工程学院

Author(s):
HUANG Wei1CAO Yujian2XU Guoming23
1. The 27th Research Institute of CETC;
2. Army Artillery and Air Defense Forces Academy of PLA;
3. Information Engineering College, Anhui Xinhua University

关键词:
深度学习偏振高光谱图像目标检测无人机
Keywords:
deep learningpolarized hyperspectral imageobject detectionunmanned aerial vehicle (UAV)
分类号:
TP391,TN911.73
DOI:
10.11846/j.issn.1001_8891.201907002
文献标志码:
A
摘要:
随着无人机等低空平台在侦察领域的不断扩展以及对性能要求的不断提高,各应用场景对目标检测精度和速度也提出了越来越高的要求。传统的目标成像方法难以满足图像质量需求,人工识别目标的方法也无法应对战场环境的快速变化。结合深度学习和偏振高光谱成像技术的发展,通过模拟偏振高光谱低空目标检测平台,提出基于Faster R-CNN的地面军事目标检测方法。采用区域建议网络模块进行模型训练,而在目标检测阶段通过对特征图进行兴趣区域池化操作得到建议特征图,最后利用建议特征图完成目标类别判定。实验选取3种典型的军事车辆缩比模型,通过偏振高光谱相机在室内外模拟环境中获取目标在不同场景条件的图像数据,以及某型无人机在低空条件下的地面车辆目标数据进行实验验证。实验表明,该方法在有效完成地面目标的检测时,能够达到理想的检测精度和速度。
Abstract:
The use of unmanned aerial vehicles (UAV) for reconnaissance requires continuous improvements in performance, as each new type of scene observed can place more stringent requirements for the accuracy and speed of object detection. Traditional object-imaging methods have difficulty meeting such requirements, and artificial object recognition is not suited for rapidly changing battle field environments. Leveraging the concomitant development of deep learning and hyperspectral polarization imaging, ground object detection based on fast R-CNNs is proposed that’s imulates a polarized hyperspectral low-altitude object detection platform. We describe a region proposal network module for training models. In its object detection phase, this approach generates a feature map by pooling feature regions into the map, which is then used to complete the object categorization decision. Three typical scaled models of military vehicles were selected to test the technique experimentally. With a polarization hyperspectral camera, object images in different scene conditions were acquired in simulated indoor and outdoor environments, and ground vehicles were successfully observed by a low-altitude UAV. The experimental results show that the proposed method achieves the ideal detection accuracy and speed when the ground object is effectively detected.

参考文献/References:

[1]? Clarke D, Grainger J F. Polarized Light and Optical Measurement[M]. Oxford: Pergamon Press, 1971: 56-60.
[2]? 杨之文. 偏振光谱的测量及研究进展[J]. 光谱实验室, 2003, 20(6): 815-820.
YANG Zhiwen. Measurement and Achievement of Polarimetric Spectrum[J]. Chinese Journal of Spectroscopy Laboratory, 2003, 20(6): 815-820.
[3]? 惠国保. 基于深层神经网络的军事目标图像分类技术[J]. 现代导航, 2016, 7(6): 430-436.
HUI Guobao. Auto target classification technology based on deep CNN[J]. Modern Navigation, 2016, 7(6): 430-436.
[4]? Hinton G E, Salakhutdinov R R. Reducing the dimensionality of data with neural networks[J]. Science, 2006, 313(5786): 504-7.
[5]? Krizhevsky A, Sutskever I, Hinton G E. ImageNet classification with deep convolutional neural networks[C]//International Conference on Neural Information Processing Systems. Curran Associates Inc. 2012: 1097-1105.
[6]? ZHANG L, ZHANG D, TIAN F. SVM and ELM: Who Wins? Object Recognition with Deep Convolutional Features from ImageNet[C]// Proceedings of ELM-2015, 2015: 249-263.?
[7]? Hosang J, Benenson R, Dollár P, et al. What Makes for Effective Detection Proposals?[J]. IEEE Transactions on Pattern Analysis & Machine Intelligence, 2016, 38(4): 814-830.
[8]? Girshick R, Donahue J, Darrell T, et al. Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation[C]//IEEE Conference on Computer Vision and Pattern Recognition, 2014: 580-587.
[9]? HE K, ZHANG X, REN S, et al. Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition[J]. IEEE Transactions on Pattern Analysis & Machine Intelligence, 2015, 37(9): 1904.
[10]? Girshick R. Fast R-CNN[J/OL]. Computer Science, 2015, https://arxiv. org/abs/1504.08083.
[11]? Vapnik V. Statistical Learning Theory[M]. John Wiley: New York, 1998.
[12]? Ren S, He K, Girshick R, et al. Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks[J]. IEEE Transactions on Pattern Analysis & Machine Intelligence, 2015, 39(6): 1137.
[13]? Simonyan K, Zisserman A. Very Deep Convolutional Networks for Large-Scale Image Recognition[J/OL]. Computer Science, 2014, https://arxiv.org/abs/1409.1556.
[14]? Everingham M, Gool L, Williams C K, et al. The Pascal Visual Object Classes (VOC) Challenge[J]. International Journal of Computer Vision, 2010, 88(2): 303-338.
[15]? Uijlings J R, Sande K E, Gevers T, et al. Selective Search for Object Recognition[J]. International Journal of Computer Vision, 2013, 104(2): 154-171.
[16]? Long J, Shelhamer E, Darrell T. Fully convolutional networks for semantic segmentation[C]//Computer Vision and Pattern Recognition. IEEE, 2015: 3431-3440.
[17]? Zeiler M D, Fergus R. Visualizing and understanding convolutional networks[C]//European Conference on Computer Vision, 2013, 8689: 818-833.

相似文献/References:

[1]王鹏翔,郭敬滨,谭文斌,等.结合多模板的多域卷积神经网络视觉跟踪算法[J].红外技术,2018,40(1):047.[doi:10.11846/j.issn.1001_8891.201801009]
 WANG Pengxiang,GUO JingbinAN Wenbin,LI Xingfei.A Multidomain CNN that Integrates Multiple Models in a Tree Structure for Visual Tracking[J].Infrared Technology,2018,40(7):047.[doi:10.11846/j.issn.1001_8891.201801009]
[2]蔡坤琪.快速图像超分辨率方法研究[J].红外技术,2018,40(3):269.[doi:10.11846/j.issn.1001_8891.201803012]
 CAI Kunqi.A Study on Rapid Image Super-resolution[J].Infrared Technology,2018,40(7):269.[doi:10.11846/j.issn.1001_8891.201803012]
[3]卢贺洋,苏胜君,袁明辉,等.太赫兹图像的超分辨率重建[J].红外技术,2019,41(1):059.[doi:10.11846/j.issn.1001_8891.201901009]
 LU Heyang,SU Shengjun,YUAN Minghui,et al.Super-resolution Reconstruction of Terahertz Images[J].Infrared Technology,2019,41(7):059.[doi:10.11846/j.issn.1001_8891.201901009]
[4]焦安波,何淼,罗海波.一种改进的HED网络及其在边缘检测中的应用[J].红外技术,2019,41(1):072.[doi:10.11846/j.issn.1001_8891.201901011]
 JIAO Anbo,HE Miao,LUO Haibo.Research on Significant Edge Detection of Infrared Image Based on Deep Learning[J].Infrared Technology,2019,41(7):072.[doi:10.11846/j.issn.1001_8891.201901011]

备注/Memo

备注/Memo:
收稿日期:2017-08-23;修订日期:2019-05-14.
作者简介:黄伟(1975-),男,高级工程师,主要研究方向为激光成像技术及应用。E-mail:89668550@qq.com。
通信作者:徐国明(1979-),男,博士后,副教授,主要研究领域为计算机视觉、图像稀疏表示及超分辨率重建。E-mail:xgm121@163.com。
基金项目:国家自然科学基金项目(61379105);中国博士后科学基金项目(2016M592961);安徽省自然科学基金项目(1608085MF140)。

更新日期/Last Update: 2019-07-12