[1]齐 航,袁健全,李 磊,等.基于深度学习的红外烟幕区域分割技术[J].控制与信息技术(原大功率变流技术),2019,(04):18-22.[doi:10.13889/j.issn.2096-5427.2019.04.400]
 QI Hang,YUAN Jianquan,LI Lei,et al.A Method of Smoke Area Segmentation for Infrared Images Based on Deep Learning[J].High Power Converter Technology,2019,(04):18-22.[doi:10.13889/j.issn.2096-5427.2019.04.400]
点击复制

基于深度学习的红外烟幕区域分割技术()
分享到:

《控制与信息技术》(原《大功率变流技术》)[ISSN:2095-3631/CN:43-1486/U]

卷:
期数:
2019年04期
页码:
18-22
栏目:
“中国飞行力学学术年会”专刊
出版日期:
2019-08-05

文章信息/Info

Title:
A Method of Smoke Area Segmentation for Infrared Images Based on Deep Learning
文章编号:
2096-5427(2019)04-0018-05
作者:
齐 航1袁健全1 李 磊1任 君2 梁 杰2
(1. 复杂系统控制与智能协同技术重点实验室,北京 100074;2.北京机电工程研究所,北京 100074)
Author(s):
QI Hang1 YUAN Jianquan1 LI Lei1REN Jun2LIANG Jie2
( 1.Science and Technology on Complex System Control and Intelligent Agent Cooperation Laboratory, Beijing 100074, China; 2. Institute of Mechanical and Electrical Engineering, Beijing 100074, China )
关键词:
深度学习图像语义分割红外图像粒子系统图像增广
Keywords:
deep learning image semantic segmentation infrared image particle system data augmentation
分类号:
TP391
DOI:
10.13889/j.issn.2096-5427.2019.04.400
文献标志码:
A
摘要:
红外成像制导系统容易受到烟幕干扰,分割烟幕区域是提高红外成像自动目标识别性能的有效技术途径。文章提出一种基于深度学习的红外烟幕干扰分割技术,针对实拍红外烟幕图像样本获取成本高而难以满足算法训练需求的问题,提出基于粒子系统的红外烟幕仿真技术,并针对性地借助图像增广技术模拟不同成像条件,借助Deeplab v3+算法实现烟幕区域分割。文章在零实拍红外图像条件下训练网络,在实拍图像集上测试结果与真实烟幕区域的平均交并比达到79%,实现了红外图像烟幕区域分割。
Abstract:
Smoke area segmentation is an efficient way to improve the performance of the infrared imaging guidance system, even though smoke is one of the main disturbances for automatic target recognition. In this paper, we proposed a smoke area segmentation method for infrared images based on deep learning. In order to reduce the cost of data acquisition, the deep learning network for cloud segmentation is trained on our simulated image data set up by particle system and data augmentation, which substantially decreases the demand for real smoke images. Then, the Deeplab v3+ algorithm was used for smoke segmentation. When testing on the real image data set, a satisfactory segmentation result with 79 percent has been reached which can meet the usage requirements, illustrating that our method is effective and efficient.

参考文献/References:

[1] LONG J, SHELHAMER E, DARRELL T. Fully Convolutional Networks for Semantic Segmentation[J]. IEEE Transactions on Pattern Analysis & Machine Intelligence, 2014, 39(4):640-651.
 [2] RONNEBERGER O, FISCHER P, BROX T. U-Net: Convolutional Networks for Biomedical Image Segmentation[C]// International Conference on Medical Image Computing & Computer-assisted Intervention, 2015.
 [3] CHEN L C, PAPANDREOU G, KOKKINOS I, et al. Semantic Image Segmentation with Deep Convolutional Nets and Fully Connected CRFs[J]. Computer Science, 2014(4):357-361.
[4] CHEN L C, ZHU Y, PAPANDREOU G, et al. Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation [J/OL]. arXiv, 2018:1-17[2018-08-22]. https://arxiv.org/pdf/1802.02611.pdf.
 [5] 王文朋, 毛文涛, 何建樑,等. 基于深度迁移学习的烟雾识别方法[J].计算机应用, 2017, 37(11): 3176-3181.
[6] 陈洋, 范荣双, 王竞雪,等. 基于深度学习的资源三号卫星遥感影像云检测方法[J].光学学报, 2018(1): 354-359.
[7] 杨永国. 基于OSG 粒子系统的飞行器发射特效的技术实现[J].舰船电子工程, 2015(3): 46-48.
[8] 侯学隆, 宋伟健. OSG的粒子特效仿真[J]. 电脑编程技巧与维护, 2010(11):65-70.
 [9] CHEN L C, PAPANDREOU G, KOKKINOS I, et al. DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs[J].IEEE Transactions on Pattern Analysis & Machine Intelligence, 2018, 40(4):834-848.
[10] CHEN L C, PAPANDREOU G, SCHROFF F, et al. Rethinking Atrous Convolution for Semantic Image Segmentation[J/OL]. arXiv, 2017:1-14[2017-12-05]. https://arxiv.org/pdf/1706.05587v3.pdf.

相似文献/References:

[1]熊群芳,林 军,刘 悦,等.深度学习研究现状及其在轨道交通领域的应用[J].控制与信息技术(原大功率变流技术),2018,(02):1.[doi:10.13889/j.issn.2096-5427.2018.02.001]
 XIONG Qunfang,LIN Jun,LIU Yue,et al.Deep Learning and Its Application in the Field of Rail Transit[J].High Power Converter Technology,2018,(04):1.[doi:10.13889/j.issn.2096-5427.2018.02.001]
[2]熊群芳,林 军,岳 伟,等. 基于深度学习的疲劳驾驶状态检测方法[J].控制与信息技术(原大功率变流技术),2018,(06):1.[doi:10.13889/j.issn.2096-5427.2018.06.400]
 XIONG Qunfang,LIN Jun,YUE Wei,et al. A Method of Fatigue Driving State Detection Based on Deep Learning[J].High Power Converter Technology,2018,(04):1.[doi:10.13889/j.issn.2096-5427.2018.06.400]
[3]熊群芳,林 军,岳 伟.基于深度学习的疲劳驾驶状态检测方法[J].控制与信息技术(原大功率变流技术),2018,(06):91.[doi:10.13889/j.issn.2096-5427.2018.06.400]
 XIONG Qunfang,LIN Jun,YUE Wei.A Method of Fatigue Driving State Detection Based on Deep Learning[J].High Power Converter Technology,2018,(04):91.[doi:10.13889/j.issn.2096-5427.2018.06.400]
[4]丁 驰,林 军,游 俊,等.基于深度学习的手势识别方法[J].控制与信息技术(原大功率变流技术),2018,(06):96.[doi:10.13889/j.issn.2096-5427.2018.06.016]
 DING Chi,LIN Jun,YOU Jun,et al.A Gesture Recognition Method Based on Deep Learning[J].High Power Converter Technology,2018,(04):96.[doi:10.13889/j.issn.2096-5427.2018.06.016]
[5]刘 悦,林 军,游 俊.语音识别技术在车载领域的应用及发展[J].控制与信息技术(原大功率变流技术),2019,(02):1.[doi:10.13889/j.issn.2096-5427.2019.02.001]
 LIU Yue,LIN Jun,YOU Jun.Application and Development of Automatic Speech Recognition in Vehicle Field[J].High Power Converter Technology,2019,(04):1.[doi:10.13889/j.issn.2096-5427.2019.02.001]
[6]高 群,朱 均,王芊芊,等. 基于鱼眼图像的目标检测算法研究[J].控制与信息技术(原大功率变流技术),2019,(03):1.[doi:10.13889/j.issn.2096-5427.2019.03.100]
 GAO Qun,ZHU Jun,WANG Qianqian,et al.Research on the Object Detection Algorithm Based on Fisheye Image[J].High Power Converter Technology,2019,(04):1.[doi:10.13889/j.issn.2096-5427.2019.03.100]
[7]高 群,朱 均,王芊芊,等.基于鱼眼图像的目标检测算法研究[J].控制与信息技术(原大功率变流技术),2019,(03):43.[doi:10.13889/j.issn.2096-5427.2019.03.100]
 GAO Qun,ZHU Jun,WANG Qianqian,et al.Research on the Object Detection Algorithm Based on Fisheye Image[J].High Power Converter Technology,2019,(04):43.[doi:10.13889/j.issn.2096-5427.2019.03.100]
[8]齐 航,袁健全,李 磊,等. 基于深度学习的红外烟幕区域分割技术[J].控制与信息技术(原大功率变流技术),2019,(04):1.[doi:10.13889/j.issn.2096-5427.2019.04.400]
 QI Hang,YUAN Jianquan,LI Lei,et al. A Method of Smoke Area Segmentation for Infrared Images Based on Deep Learning[J].High Power Converter Technology,2019,(04):1.[doi:10.13889/j.issn.2096-5427.2019.04.400]

备注/Memo

备注/Memo:
收稿日期:2019-05-15
作者简介:齐航(1995—),女,硕士研究生,主要研究方向为导航、制导与控制;袁健全(1963—),男,研究员,主要研究方向为导航、制导与控制。
基金项目:国家自然科学基金(61803356);国防基础科研计划(JCKY 2017204B064)
更新日期/Last Update: 2019-08-20