复杂背景的夜光遥感建筑区检测
作者:
作者单位:

1.华中科技大学 人工智能与自动化学院 多谱信息处理技术国家级重点实验室,湖北 武汉 430074;2.上海交通大学 机器人研究所,上海 200240

中图分类号:

TP753

基金项目:

国家自然科学基金 (61773389)


Detection of building area with complex background by night light remote sensing
Author:
Affiliation:

1.National Key Laboratory of Multi-spectral Information Processing Technology, Huazhong University of Science and Technology, Wuhan 430074, China;2.Institute of Robotics, Shanghai Jiaotong University, Shanghai 200240, China

Fund Project:

Supported by the National Natural Science Foundation of China (61773389)

  • 摘要
  • | |
  • 访问统计
  • |
  • 参考文献 [30]
  • |
  • 相似文献 [20]
  • | | |
  • 文章评论
    摘要:

    提出了一种新的解决夜光遥感复杂背景问题的单阶段深度卷积检测网络,首先通过提取高维特征再特征选择的思想设计分类网络提取语义特征,并研究不同的通道数网络对降噪的影响;提出灰度能量的先验框匹配,将低噪声高质量的匹配框输入SSD检测网络,并使用积分图思想简化计算;使用可变形卷积以适应目标的形变,并获取更强的几何特征表达能力;通过加入顺序连接与密集连接改进全局语义模块,引入了网络的跨层信息交互,其注意力图综合考虑了高低感受野以有效区分小型目标和背景噪声。在夜光遥感数据集上通过实验验证了所设计的网络相比于其他单阶段网络具有优势,对于复杂背景下的建筑区具有较好的检测效果。

    Abstract:

    A new single-stage deep convolution detection network is proposed to solve the complex background problem of night light remote sensing. Firstly, a classification network is designed by extracting high-dimensional features and then selecting features, and the influence of different channel number networks of noise reduction is studied. A prior box matching of gray-scale energy is proposed, inputting a low-noise and high-quality matching box into SSD detection network, and the idea of integral diagram is used to simplify the calculation. By adding sequential connection and dense connection to improve the global semantic module, the cross layered information interaction of the network is introduced, and its attention map comprehensively considers the high and low receptive fields to effectively distinguish small targets and background noise. Experimental results of the night light remote sensing data set show that the designed network has advantages over the rest single-stage network, which has a better detection effect of the building area under the complex background.

    参考文献
    [1] Wu Bin,Yu Bailang,Yao Shenjun, et al. A surface network based method for studying urban hierarchies by night time light remote sensing data[J]. International Journal of Geographical Information Science, 2019, 33(7). 10.1080/13658816.2019.1585540
    [2] Zhu Hui, Zhang Qingling, Zhang Shan. Temporal and spatial characteristics of social and economic development in Central Asia based on night light remote sensing from 1992 to 2017 [J]. Journal of Earth Information Science朱惠,张清凌,张珊. 1992—2017年基于夜光遥感的中亚社会经济发展时空特征分析.地球信息科学学报, 2020, 22(07): 1449-1462.
    [3] Cheng X, Shao H, Li Y, et al. Urban land intensive use evaluation study based on nighttime light—a case study of the yangtze river economic belt[J]. Sustainability, 2019, 11(3): 675. 10.3390/su11030675
    [4] Sun Lishuang, Han Yaohui, Xie Zhiwei, et al. Neighborhood extremum method for extracting urban built-up area using noctilucent remote sensing data [J]. Journal of Wuhan University (Information Science Edition孙立双,韩耀辉,谢志伟,等. 采用夜光遥感数据提取城市建成区的邻域极值法.武汉大学学报(信息科学版)): 1-7.
    [5] Li Xi, Xue Xiangyu. Estimation method of remote sensing power consumption based on Boston matrix [J]. Journal of Wuhan University Information Science Edition(李熙,薛翔宇. 基于波士顿矩阵的夜光遥感电力消费估算方法.武汉大学学报(信息科学版)), 2018, 43(12): 1994-2002.
    [6] Li Deren, Li Xi. On luminous remote sensing data mining [J]. Acta Sinica Sinica(李德仁,李熙. 论夜光遥感数据挖掘. 测绘学报), 2015, 44(06): 591-601.
    [7] Chang Y, Wang S, Zhou Y, et al. A novel method of evaluating highway traffic prosperity based on nighttime light remote sensing[J]. Remote Sensing. 2019, 12(1): 102. 10.3390/rs12010102
    [8] Jiang W, He G, Leng W, et al. Characterizing light pollution trends across protected areas in china using nighttime light remote sensing data[J]. ISPRS International Journal of Geo-Information. 2018, 7(7): 243. 10.3390/ijgi7070243
    [9] Wen Kai. Infrared dim small target detection and tracking algorithm based on complex fusion features and gray texture histogram descriptors [J]. Science, technology and engineering(闻凯. 基于复杂融合特征与灰度-纹理直方图描述子的红外弱小目标检测追踪算法. 科学技术与工程), 2016, 16(34): 83-91.
    [10] Yu B , Lian T , Huang Y , et al. Integration of nighttime light remote sensing images and taxi GPS tracking data for population surface enhancement[J]. International Journal of Geographical Information Science, 2019, 33(3-4):687-706.
    [11] Faouzi B, Washaya P. Tracking Dynamic changes and monitoring socioeconomic parameters in algeria between 1993 and 2012,using nighttime light remote sensing [J]. Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci. 2017, XLII-2/W7: 1127-1135. 10.5194/isprs-archives-xlii-2-w7-1127-2017
    [12] Devkota B, Miyazaki H, Witayangkurn A, et al. Using volunteered geographic information and nighttime light remote sensing data to identify tourism areas of interest[J]. Sustainability. 2019, 11(17): 4718. 10.3390/su11174718
    [13] Swathi R, Srinivas A. An improved image registration method using e-sift feature descriptor with hybrid optimization algorithm[J]. Journal of the Indian Society of Remote Sensing. 2020, 48(2): 215-226. 10.1007/s12524-019-01063-w
    [14] Wang Xing, Zhou Dong, Wang Wenmao. Estimation of the building density of the main urban area of hefei using luojia no.1 luminous image [J]. Remote sensing information(王兴,周侗,王文懋. 合肥主城区建筑密度的珞珈一号夜光影像估算. 遥感信息), 2020, 35(03): 71-77.
    [15] Zhang Q, Schaaf C, Seto K C. The vegetation adjusted ntl urban index: a new approach to reduce saturation and increase variation in nighttime luminosity[J]. Remote Sensing of Environment. 2013, 129. 10.1016/j.rse.2012.10.022
    [16] Zhang Guoliang, Zhu Ruifei, Du Yibo, et al. Application of jilin-1 high resolution night light remote sensing image in urban monitoring [J]. Satellite application(张国亮,朱瑞飞,杜一博,等. 吉林一号高分辨率夜光遥感影像在城市监测中的应用. 卫星应用), 2020(03): 27-33.
    [17] Yongling Yao Y L S O. House vacancy at urban areas in china with nocturnal light data of dmsp-ols[C]. o ZhIEEE International Conference on Spatial Data Mining and Geographical Knowledge Services: 2011. 10.1109/icsdm.2011.5969087
    [18] Wu Shuangchen, Zuo Zhengrong. Infrared small target detection based on deep convolution neural network [J]. J.Infrared Millim.Waves(吴双忱,左峥嵘. 基于深度卷积神经网络的红外小目标检测. 红外与毫米波学报), 2019, 38(03): 371-380.
    [19] Redmon J, Farhadi A. Yolov3: an incremental improvement[J]. arXiv e-prints. 2018: 1804-2767. 10.1109/cvpr.2018.00426
    [20] Liu W, Anguelov D, Erhan D, et al. Ssd: single shot multibox detector[C]. Cham: Springer International Publishing, 2016. 10.1007/978-3-319-46448-0_2
    [21] Tian Z, Shen C, Chen H, et al. Fcos: fully convolutional one-Stage object detection[J]. arXiv e-prints. 2019: 1355-1904. 10.1109/iccv.2019.00972
    [22] Shaoqing R, Kaiming H, Ross G, et al. Faster r-cnn: towards real-time object detection with region proposal networks.[J]. IEEE transactions on pattern analysis and machine intelligence. 2017, 39(6). 10.1109/tpami.2016.2577031
    [23] Simonyan K, Vedaldi A, Zisserman A. Deep inside convolutional networks: visualising image classification models and saliency maps[J]. arXiv e-prints. 2013: 1312-6034. 10.5244/c.27.8
    [24] Dai J, Qi H, Xiong Y, et al. Deformable convolutional networks[J]. arXiv e-prints. 2017: 1703-6211. 10.1109/iccv.2017.89
    [25] Yu F, Koltun V. Multi-scale context aggregation by dilated convolutions[J]. arXiv e-prints. 2015: 1511-7122.
    [26] Cao Y, Xu J, Lin S, et al. Gcnet: non-local networks meet squeeze-excitation networks and beyond[J]. arXiv e-prints. 2019: 1904-11492. 10.1109/iccvw.2019.00246
    [27] Changyong R, Duho P, Kyongchol R, et al. Semantic image description and classification based on generalized set [J]. ICTACT Journal on Image and Video Processing. 2018, 8(4). 10.21917/ijivp.2018.0250
    [28] Li Deren, Zhang Guo, Shen Xin, et al. Design and processing of noctilucent remote sensing of luojia-1 satellite [J]. Acta remote sensing Sinica(李德仁,张过,沈欣,等. 珞珈一号01星夜光遥感设计与处理. 遥感学报), 2019, 23(06): 1011-1022.
    [29] Liang Ze, Huang Jiao, Wei Feili, et al. Spatial range identification of urban agglomerations in china based on noctilucent remote sensing images and baidu poi data [J]. Geographic research(梁泽,黄姣,韦飞黎,等. 基于夜光遥感影像与百度POI数据的中国城市群空间范围识别方法. 地理研究), 2020, 39(01): 92-102.
    [30] Yang F, Li W, Li W, et al. S3od: single stage small object detector from scratch for remote sensing images[C]. Cham: Springer International Publishing, 2019. 10.1007/978-3-030-34113-8_29
    引证文献
    网友评论
    网友评论
    分享到微博
    发 布
引用本文

李海,李洋,左峥嵘.复杂背景的夜光遥感建筑区检测[J].红外与毫米波学报,2021,40(3):369~380]. LI Hai, LI Yang, ZUO Zheng-Rong. Detection of building area with complex background by night light remote sensing[J]. J. Infrared Millim. Waves,2021,40(3):369~380.]

复制
分享
文章指标
  • 点击次数:802
  • 下载次数: 2809
  • HTML阅读次数: 618
  • 引用次数: 0
历史
  • 收稿日期:2020-04-05
  • 最后修改日期:2021-05-12
  • 录用日期:2020-08-24
  • 在线发布日期: 2021-05-12
  • 出版日期: 2021-06-25
文章二维码