首页 > 论文 > 激光与光电子学进展 > 57卷 > 2期(pp:21507--1)


Real-Time Pedestrian Detection for Far-Infrared Vehicle Images and Adaptive Instance Segmentation

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


针对红外图像检测与分割任务中颜色信息缺失,特征细节模糊并带有噪声,当目标数量较多时传统方法提取过程速度较慢等问题,提出一种用于远红外图像的优化YOLO检测与分割网络模型。提出的两个优化点分别为:综合分析实验使用的两种远红外数据集后使用K-means++聚类算法寻找多尺度预测标记锚点框尺寸;使用局部检测位置自适应阈值分割方法对检测目标进行像素级实例分割。本文优化算法在FLIR公开数据集与本文数据集中的检测速度分别为29 frame/s与28 frame/s,保证了实时输出的要求;行人检测准确率分别达到75.3%与77.6%,分割结果平均交并比达到70%~90%。实验结果表明,本文算法具有良好的稳健性和普适性,在远红外图像中可快速有效地检测行人并生成实例掩模。


In infrared image detection and segmentation tasks, the color information is lost, the features are fuzzy with noise, the target number is large, and the traditional extraction method is slow. Therefore, we propose an optimized YOLO detection and segmentation network model for far-infrared images. The two proposed optimization points are as follows. We use the K-means++ clustering algorithm to determine the multi-scale prediction anchor size after the analysis of two far-infrared databases. We also perform pixel-level instance segmentation of detection targets using localized adaptive threshold segmentation. The experimental results show that the proposed algorithm performs pedestrian detection at detection speeds of 29 frame/s and 28 frame/s on the FLIR dataset and the dataset used in this paper, respectively, ensuring the requirement of real-time output. The pedestrian detection accuracies in these datasets reach 75.3% and 77.6%. Moreover, the average intersection over the union of the segmentation results is 70%--90%. In summary, the algorithm performs well with respect to robustness and universality. The algorithm provides a valuable reference method for pedestrian detection and segmentation in far-infrared fields.









作者单位    点击查看

于博:大连海事大学信息科学技术学院, 辽宁 大连 116026
马书浩:大连海事大学信息科学技术学院, 辽宁 大连 116026
李红艳:大连海事大学信息科学技术学院, 辽宁 大连 116026
李春庚:大连海事大学信息科学技术学院, 辽宁 大连 116026
安居白:大连海事大学信息科学技术学院, 辽宁 大连 116026



【1】Braga-Neto U M, Choudhury M, Goutsias J I. Automatic target detection and tracking in forward-looking infrared image sequences using morphological connected operators [J]. Journal of Electronic Imaging. 2004, 13(4): 802-813.

【2】Corsi C. Infrared: a key technology for security systems [J]. Advances in Optical Technologies. 2012, 2012: 838752.

【3】Cui M Y. Application field and technical characteristics of infrared thermal imager China Security & Protection[J]. 0, 2014(12): 90-93.
崔美玉. 论红外热像仪的应用领域及技术特点 中国安防[J]. 0, 2014(12): 90-93.

【4】Wang H X, Dong H, Zhou Z Q. Review on dim small target detection technologies in infrared single frame images [J]. Laser & Optoelectronics Progress. 2019, 56(8): 080001.
王好贤, 董衡, 周志权. 红外单帧图像弱小目标检测技术综述 [J]. 激光与光电子学进展. 2019, 56(8): 080001.

【5】Liu R, Wang D J, Jia P, et al. Overview on small target detection technology in infrared image [J]. Laser & Optoelectronics Progress. 2016, 53(5): 050004.
刘让, 王德江, 贾平, 等. 红外图像弱小目标探测技术综述 [J]. 激光与光电子学进展. 2016, 53(5): 050004.

【6】Peng Z Y, Wang X J, Lu J. Infrared target detection under hot dome infrared radiation based on visual saliency method [J]. Infrared and Laser Engineering. 2014, 43(6): 1772-1776.
彭志勇, 王向军, 卢进. 窗口热辐射下基于视觉显著性的红外目标检测方法 [J]. 红外与激光工程. 2014, 43(6): 1772-1776.

【7】Cai W J, Wang L P, Zhang L P. Vehicle detection algorithm based on SLPP-SHOG in infrared image [J]. Laser & Infrared. 2016, 46(8): 1018-1022.
蔡文靖, 王鲁平, 张路平. 基于SLPP-SHOG的红外图像车辆检测方法 [J]. 激光与红外. 2016, 46(8): 1018-1022.

【8】Fan Q S, Fan H B, Lin Y, et al. Multi-object extraction methods based on long-line column scanning for infrared panorama imaging [J]. Infrared Technology. 2019, 41(2): 118-126.
范青帅, 范宏波, 林宇, 等. 基于长线列扫描周视红外成像的多目标提取方法综述 [J]. 红外技术. 2019, 41(2): 118-126.

【9】Chen Q, Sheng H X, Zhang Z, et al. Moving object detection under infrared light mutation [J]. Laser & Optoelectronics Progress. 2016, 53(11): 111005.
陈强, 盛惠兴, 张卓, 等. 红外光照突变下的运动目标检测 [J]. 激光与光电子学进展. 2016, 53(11): 111005.

【10】Bertozzi M, Broggi A, Fascioli A, et al. Pedestrian detection for driver assistance using multiresolution infrared vision [J]. IEEE Transactions on Vehicular Technology. 2004, 53(6): 1666-1678.

【11】Zhang L, Wu B, Nevatia R. Pedestrian detection in infrared images based on local shape features . [C]∥2007 IEEE Conference on Computer Vision and Pattern Recognition, June 17-22, 2007, Minneapolis, MN, USA. New York: IEEE. 2007, 9738285.

【12】Suard F, Rakotomamonjy A, Bensrhair A, et al. Pedestrian detection using infrared images and histograms of oriented gradients . [C]∥2006 IEEE Intelligent Vehicles Symposium, June 13-15, 2006, Meguro-Ku, Japan. New York: IEEE. 2006, 206-212.

【13】Ge J F, Luo Y P, Tei G. Real-time pedestrian detection and tracking at nighttime for driver-assistance systems [J]. IEEE Transactions on Intelligent Transportation Systems. 2009, 10(2): 283-298.

【14】Ma Y, Chang Q, Hu M F. Research on infrared human detection from complex backgrounds [J]. Infrared Technology. 2017, 39(11): 1038-1044, 1053.
马也, 常青, 胡谋法. 复杂背景下红外人体目标检测算法研究 [J]. 红外技术. 2017, 39(11): 1038-1044, 1053.

【15】Liu F, Wang S B, Wang X J, et al. Infrared pedestrian detection method in low visibility environment based on multi feature association [J]. Infrared and Laser Engineering. 2018, 47(6): 0604001.
刘峰, 王思博, 王向军, 等. 多特征级联的低能见度环境红外行人检测方法 [J]. 红外与激光工程. 2018, 47(6): 0604001.

【16】Su X Q, Sun S Y, Ge M, et al. Pedestrian detection and tracking of vehicle infrared images [J]. Laser & Infrared. 2012, 42(8): 949-953.
苏晓倩, 孙韶媛, 戈曼, 等. 车载红外图像的行人检测与跟踪技术 [J]. 激光与红外. 2012, 42(8): 949-953.

【17】Che K, Xiang Z T, Chen Y F, et al. Research on infrared image pedestrian detection based on improved fast R-CNN [J]. Infrared Technology. 2018, 40(6): 578-584.
车凯, 向郑涛, 陈宇峰, 等. 基于改进Fast R-CNN的红外图像行人检测研究 [J]. 红外技术. 2018, 40(6): 578-584.

【18】Xu M, Yu X S, Chen D Y, et al. Pedestrian detection in complex thermal infrared surveillance scene [J]. Journal of Image and Graphics. 2018, 23(12): 1829-1837.
许茗, 于晓升, 陈东岳, 等. 复杂热红外监控场景下行人检测 [J]. 中国图象图形学报. 2018, 23(12): 1829-1837.

【19】Girshick R. Fast R-CNN . [C]∥2015 IEEE International Conference on Computer Vision (ICCV), December 7-13, 2015, Santiago, Chile. New York: IEEE. 2015, 1440-1448.

【20】Ren S Q, He K M, Girshick R, 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): 1137-1149.

【21】Liu W, Anguelov D, Erhan D, et al. SSD: single shot MultiBox detector [M]. ∥Leibe B, Matas J, Sebe N, et al. Computer vision-ECCV 2016. Lecture notes in computer science. Cham: Springer. 2016, 9905: 21-37.

【22】Redmon J, Divvala S, Girshick R, et al. You only look once: unified, real-time object detection . [C]∥Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, June 26-July 1, 2016, Las Vegas, Nevada. New York: IEEE. 2016, 779-788.

【23】Redmon J, Farhadi A. YOLO9000: better, faster, stronger . [C]∥Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, July 21-26, 2017, Honolulu, Hawaii, USA. New York: IEEE. 2017, 7263-7271.

【24】Redmon J. -04-08)[2019-04-16] . https:∥arxiv.gg363. 2018, site/abs/1804: 02767.

【25】Hartigan J A, Wong M A. Algorithm AS 136: a K-means clustering algorithm [J]. Journal of the Royal Statistical Society. Series C (Applied Statistics). 1979, 28(1): 100-108.

【26】Arthur D, Vassilvitskii S. K-means++: the advantages of careful seeding . [C]∥Proceedings of the eighteenth annual ACM-SIAM symposium on Discrete algorithms, January 7-9, 2007, New Orleans, Louisiana. USA: Society for Industrial and Applied Mathematics Philadelphia. 2007, 1027-1035.

【27】Otsu N. A threshold selection method from gray-level histograms [J]. IEEE Transactions on Systems, Man, and Cybernetics. 1979, 9(1): 62-66.


Yu Bo,Ma Shuhao,Li Hongyan,Li Chungeng,An Jubai. Real-Time Pedestrian Detection for Far-Infrared Vehicle Images and Adaptive Instance Segmentation[J]. Laser & Optoelectronics Progress, 2020, 57(2): 021507

于博,马书浩,李红艳,李春庚,安居白. 远红外车载图像实时行人检测与自适应实例分割[J]. 激光与光电子学进展, 2020, 57(2): 021507

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