首页 > 论文 > 红外 > 37卷 > 12期(pp:1-9)

平面地标导引下的红外建筑物目标间接识别技术研究

Research on Indirect Recognition Technology of Infrared Buildings under Guidance of Planar Landmarks

  • 摘要
  • 论文信息
  • 参考文献
  • 被引情况
  • PDF全文
分享:

摘要

在某些情况下,如目标处于远距离、目标 局部对比度低、目标被遮挡、目标区模式频繁出现等,采用 直接识别方法难以正确识别目标。为此,提出了一种基 于平面地标间接定位的典型地面建筑物目标识别方法。在地面准 备阶段进行地标选取以及参考图制备。在飞行阶段,用匹配算法 对地标进行自动捕获、识别和定位,然后利用地标与目标的空间约束关系,间接 定位包含目标的局部感兴趣区。当局部感兴趣区的对比度显著 时,启动最后阶段的基于形态学处理框架的直接识别操作。通 过仿真实验测试分析了该算法的适应性。

Abstract

In some cases, for instance, when a target is located at a long distance, when a target has low local contrast, when a target is obscured and when the pattern of a target zone occurs frequently, it is difficult for a direct recognition method to correctly recognize the target. For this reason, a recognition method for typical ground buildings based on indirect location of planar landmarks is proposed. At the ground preparation stage, the landmarks are selected and their reference images are prepared. At the flight stage, the landmarks are captured, recognized and located by a matching algorithm automatically. Then, the spatial constrains between the target and the landmarks are used to indirectly locate the local region of interest which contains the target. When the local region of interest has high contrast, the direct recognition based on morphological processing framework at final stage is started. The adaptability of the proposed algorithm is tested and analyzed in the simulation experiment.

Newport宣传-MKS新实验室计划
补充资料

中图分类号:TP391.4

DOI:10.3969/j.issn.1672-8785.2016.12.001

基金项目:中央高校基本科研业务费(ZYGX2015KYQD032)

收稿日期:2016-09-06

修改稿日期:--

网络出版日期:--

作者单位    点击查看

王登位:电子科技大学航空航天学院,四川 成都 611731

联系人作者:王登位(wdengwei@126.com)

备注:王登位(1982-),男,贵州遵义人,博士,讲师,主要从事红外成像目标识别方面的研究工作。

【1】Lin C, Huertas A, Nevatia R. Detection of Buildings Using Perceptual Grouping and Shadows [C]. Seattle: IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 1994.

【2】张煜, 张祖勋, 张剑清. 几何约束与影像分割相结合的快速半自动房屋提取 [J]. 武汉测绘科技大学学报, 2000, 25(3): 238-242.

【3】张前进, 郭雷在. 基于特征学习的建筑物自动识别算法研究 [J]. 弹箭与制导学报, 2007, 27(3): 199-202.

【4】陶文兵, 柳健, 田金文. 一种新型的航空图像城区建筑物自动提取方法 [J]. 计算机学报, 2003, 26(7): 867-973.

【5】金泰松, 叶聪颖, 李翠华, 等. 一种复杂场景下建筑目标识别方法 [J]. 计算机工程, 2007, 33(6): 198-200.

【6】Zhao C R, Liu C C, Lai Z H. Multi-scale Gist Feature Manifold for Building Recognition [J].Neurocomputing, 2011, 74(17): 2929-2940.

【7】Ren K Y, Sun H X, Jia Q X, et al. Building Recognition from Aerial Images Combining Segmentation and Shadow [C]. Shanghai: IEEE International Conference on Intelligent Computing and Intelligent Systems, 2009.

【8】Trinh H H, Kim D N, Jo K H. Supervised Training Database by Using SVD-based Method for Building Recognition [C]. Singapore: International Conference on Control, Automation and Systems, 2008.

【9】Shi X P, Chen R X. Building Recognition from High Resolution Image [C]. Wuhan: International Conference on Information Engineering and Computer Science, 2009.

【10】Guenther Riedi, Michael Assel, Alfred Fendt, et al. IR Focal Plane Array Seekers for Ground-to-Ground and Air-to-Ground Missiles [C].SPIE, 2001, 4369: 201-209.

【11】明宝印, 尹健, 高士英. KEPD-350空地导弹先进技术剖 析 [J]. 飞航导弹, 2006, 20(7): 4-6.

引用该论文

WANG Deng-wei. Research on Indirect Recognition Technology of Infrared Buildings under Guidance of Planar Landmarks[J]. INFRARED, 2016, 37(12): 1-9

王登位. 平面地标导引下的红外建筑物目标间接识别技术研究[J]. 红外, 2016, 37(12): 1-9

您的浏览器不支持PDF插件,请使用最新的(Chrome/Fire Fox等)浏览器.或者您还可以点击此处下载该论文PDF