留言板

尊敬的读者、作者、审稿人, 关于本刊的投稿、审稿、编辑和出版的任何问题, 您可以本页添加留言。我们将尽快给您答复。谢谢您的支持!

姓名
邮箱
手机号码
标题
留言内容
验证码

基于红外特征的三维目标识别算法研究

夏琰

夏琰. 基于红外特征的三维目标识别算法研究[J]. 红外技术, 2022, 44(11): 1161-1166.
引用本文: 夏琰. 基于红外特征的三维目标识别算法研究[J]. 红外技术, 2022, 44(11): 1161-1166.
XIA Yan. Research on 3D Target Recognition Algorithm Based on Infrared Features[J]. Infrared Technology , 2022, 44(11): 1161-1166.
Citation: XIA Yan. Research on 3D Target Recognition Algorithm Based on Infrared Features[J]. Infrared Technology , 2022, 44(11): 1161-1166.

基于红外特征的三维目标识别算法研究

基金项目: 

吉林省优秀青年人才基金项目 20210103154JH

详细信息
    作者简介:

    夏琰(1980-),女,硕士,副教授,主要研究方向为计算机应用及图像处理方面的研究。E-mail: wsxiayan99@163.com

  • 中图分类号: TP391.41

Research on 3D Target Recognition Algorithm Based on Infrared Features

  • 摘要: 基于三维特征的目标识别存在相似点云域容易误判、总数据运算量大等问题,而造成目标检出率低和误判率高。为了提高目标识别准确度与速度,提出了基于红外特征的三维目标识别算法。系统同时获取目标区域的二维红外图像与三维点云数据,利用目标红外特性的显著特征获得目标的投影范围,并计算系统与目标的位姿关系。根据红外特征映射关系计算点云数据中目标的限定范围,由此大幅缩减需要匹配计算的点云总量。在相同背景条件下对同一目标车辆进行测试,记录分析了3种不同测试角度条件下的识别数据。结果显示,传统点云识别算法的目标检出率均值为93.4%,误判率均值为19.5%,收敛耗时4.77 s。本算法的目标检出率均值为98.7%,误判率均值为1.5%,收敛耗时1.23 s。由此可见,基于红外特征的目标识别算法的检出率和误判率都更有优势,且处理速度更快。
  • 图  1  基于红外特征的三维目标识别系统

    Figure  1.  3D target recognition system based on infrared features

    图  2  激光雷达获取点云与红外图像融合的位置关系

    Figure  2.  The position relationship between point cloud acquired by lidar and infrared image fusion

    图  3  基于红外特征的识别算法程序流程

    Figure  3.  Flowchart of the recognition algorithm program based on infrared features

    图  4  目标红外图像获取与目标区域框选限定

    Figure  4.  Target infrared image acquisition and target area frame selection

    图  5  点云数据及限定后的目标点云域

    Figure  5.  Point cloud data and limited target point cloud domain

    图  6  收敛时间趋势对比

    Figure  6.  Convergence time trend comparison

    表  1  识别结果对比数据

    Table  1.   Identification results comparison

    Comparison
    group
    G1 G2 G3
    Pd/% Pf/% Pd/% Pf/% Pd/% Pf/%
    DMM 99.4 2.2 94.2 2.5 93.6 2.7
    EB 96.2 17.5 90.4 24.6 93.5 16.4
    SV 91.7 5.2 97.9 4.9 98.2 5.4
    IFTR 99.3 1.6 98.2 1.7 98.7 1.3
    下载: 导出CSV
  • [1] REN S, HE K, Girshick R, et al. Faster R-CNN: Real-time target detection in regional planning Network[J]. IEEE Pair Model Analysis and Machinery Information, 2017, 39(6): 1137-1149.
    [2] GUO Y L, Sohel F, Bennamoun M, et al. A novel local surface feature for 3D object recognition under clutter and occlusion[J]. Information Sciences, 2015, 293: 196-213. doi:  10.1016/j.ins.2014.09.015
    [3] 王果, 王成, 张振鑫, 等. 利用车载激光点云的分车带识别及单木分割方法[J]. 激光与红外, 2020, 50(11): 1333-1337. doi:  10.3969/j.issn.1001-5078.2020.11.008

    WANG Guo, WANG Cheng, ZHANG Zhenxin, et al. Single tree segmentation method of urban distributing belt based on vehicle-borne laser point cloud data[J]. Laser & Infrared, 2020, 50(11): 1333-1337. doi:  10.3969/j.issn.1001-5078.2020.11.008
    [4] 胡海瑛, 惠振阳, 李娜. 基于多基元特征向量融合的机载LiDAR点云分类[J]. 中国激光, 2020, 47(8): 229-239. https://www.cnki.com.cn/Article/CJFDTOTAL-JJZZ202008029.htm

    HU Haiying, HUI Zhenyang, LI Na. Airborne LiDAR point cloud classification based on multiple-entity eigenvector fusion[J]. Chinese Journal of Lasers, 2020, 47(8): 229-239. https://www.cnki.com.cn/Article/CJFDTOTAL-JJZZ202008029.htm
    [5] Sochor J, Spaňhel J, Herout A. Boxcars: improving fine-grained recognition of vehicles using 3-d bounding boxes in traffic surveillance[J]. IEEE Transactions on Intelligent Transportation Systems, 2018, 20(1): 97-108.
    [6] Sochor J, Herout A, Havel J. BoxCars: 3D boxes as CNN input for improved fine-grained vehicle recognition[C]// IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016: 3006-3015.
    [7] Dubská M, Herout A, Juránek R, et al. Fully automatic roadside camera calibration for traffic surveillance[J]. IEEE Transactions on Intelligent Transportation Systems, 2014, 16(3): 1162-1171.
    [8] Dubská M, Herout A, Sochor J. Automatic camera calibration for traffic understanding[C]// Proceedings of the British Machine Vision Conference(BMVC), 2014: 1-12.
    [9] GISEOK K, JAE- SOO C. Vision- Based vehicle detection and inter- vehicle distance estimation for driver alarm system[J]. Optical Review, 2012, 25(6): 388- 393.
    [10] 薛培林, 吴愿, 殷国栋, 等. 基于信息融合的城市自主车辆实时目标识别[J]. 机械工程学报, 2020, 56(12): 165-173. https://www.cnki.com.cn/Article/CJFDTOTAL-JXXB202012021.htm

    XUE Peilin, WU Yuan, YIN Guodong, et al. Real-time target recognition for urban autonomous vehicles based on information fusion[J]. Journal of Mechanical Engineering, 2020, 56(12): 165-173. https://www.cnki.com.cn/Article/CJFDTOTAL-JXXB202012021.htm
    [11] 仝选悦, 吴冉, 杨新锋, 等. 红外与激光融合目标识别方法[J]. 红外与激光工程, 2018, 47(5): 158-165. https://www.cnki.com.cn/Article/CJFDTOTAL-HWYJ201805025.htm

    TONG Xuanyue, WU Ran, YANG Xinfeng, et al. Fusion target recognition method of infrared and laser[J]. Infrared and Laser Engineering, 2018, 47(5): 158-165. https://www.cnki.com.cn/Article/CJFDTOTAL-HWYJ201805025.htm
    [12] YAN Y, MAO Y, LI B. Second: sparsely embedded convolutional detection[J]. Sensors, 2018, 18(10): 3337/1-17.
  • 加载中
图(6) / 表(1)
计量
  • 文章访问数:  36
  • HTML全文浏览量:  14
  • PDF下载量:  17
  • 被引次数: 0
出版历程
  • 收稿日期:  2022-04-12
  • 修回日期:  2022-07-28
  • 刊出日期:  2022-11-20

目录

    /

    返回文章
    返回