[1]于龙姣,于博,李春庚,等.优化卷积网络及低分辨率热成像的夜间人车检测与识别[J].红外技术,2020,42(7):651-659.[doi:10.11846/j.issn.1001_8891.202007008]
 YU Longjiao,YU Bo,LI Chungeng,et al.Detection and Recognition of Persons and Vehicles in Low-Resolution Nighttime Thermal Images Based on Optimized Convolutional Neural Network[J].Infrared Technology,2020,42(7):651-659.[doi:10.11846/j.issn.1001_8891.202007008]
点击复制

优化卷积网络及低分辨率热成像的夜间人车检测与识别
分享到:

《红外技术》[ISSN:1001-8891/CN:CN 53-1053/TN]

卷:
42卷
期数:
2020年第7期
页码:
651-659
栏目:
出版日期:
2020-07-23

文章信息/Info

Title:
Detection and Recognition of Persons and Vehicles in Low-Resolution Nighttime Thermal Images Based on Optimized Convolutional Neural Network
文章编号:
1001-8891(2020)07-0651-09
作者:
于龙姣于博李春庚安居白
大连海事大学 信息科学技术学院
Author(s):
YU LongjiaoYU BoLI ChungengAN Jubai
School of Information Science and Technology, Dalian Maritime University
关键词:
自动驾驶夜间环境人车检测与识别红外热成像Faster RCNN
Keywords:
self-driving car night time environment detection and recognition of persons & vehicles infrared thermal imaging Faster RCNN
分类号:
TP183
DOI:
10.11846/j.issn.1001_8891.202007008
文献标志码:
A
摘要:
夜间环境下人车的检测与识别在自动驾驶,安防等领域具有重要意义。本文提出使用性价比较高的低分辨率红外热成像摄像机拍摄的图像来进行夜间的人车检测与识别,并根据图像独特的性质对Faster RCNN网络进行了优化。增加多通道卷积层来适应热成像图像的灰度特性。使用全局平均池化层来适应较少的图像及类别数量,增加批标准化层来防止加深加宽网络后可能出现的梯度消失或爆炸。使用在城市夜间环境中采集的2000张低分辨率热成像图像对网络进行训练与测试,平均准确识别率达到71.3%。相比于传统的检测手段,本组合方法在真实的场景中取得了较好的识别效果,同时提升了准确识别率,有效解决了夜间环境下人车检测与识别的问题,鲁棒性及应用价值较强。
Abstract:
The detection and recognition of persons and vehicles in the nighttime environment is highly important in the fields of self-driving cars and security. This paper proposes to use images taken by a cost-effective low-resolution infrared thermal imaging camera. We optimize the faster region-based convolutional neural network according to the unique nature of the images. A multi-channel convolution layer is added to accommodate the grayscale characteristics of thermographic images. We use a global average pooling layer so that fewer images and categories are needed, and we add batch normalization layers to prevent the appearance of exploding or vanishing gradients after the network is widened. The network is trained and tested using 2000 low-resolution thermal images collected in an urban nighttime environment. The average accurate recognition rate is 71.3%, indicating that the method effectively solves the problem of detection and recognition of persons and vehicles in the nighttime environment. The stickiness value and application potential are high.

参考文献/References:

[1] Holstein T, Dodig-Crnkovic G, Pelliccione P. Ethical and Social Aspects of Self-Driving Cars[EB/OL]. Arxiv preprint, 2018, arXiv:1802.04103.
[2] Gaussorgues G, Chomet S. Infrared Thermography[M]. Springer Science & Business Media,2012, 5:379-395
[3] Beard J. Introduction to infrared thermography[J]. Lecture Presented at IT, 2007, 570: 7-14.?
[4] Meis U, Ritter W, Neumann H. Detection and classification of obstacles in night vision traffic scenes based on infrared imagery[C]//Proceedings of the 2003 IEEE International Conference on Intelligent Transportation Systems, 2003, 2: 1140-1144.
[5] Cao Y, Pranata S, Nishimura H. Local binary pattern features for pedestrian detection at night/dark environment[C]//2011 18th IEEE International Conference on Image Processing. IEEE, 2011: 2053-2056.
[6] Chen T H, Chen J L, Chen C H, et al. Vehicle detection and counting by using headlight information in the dark environment[C]//Third International Conference on Intelligent Information Hiding and Multimedia Signal Processing (IIH-MSP 2007). IEEE, 2007, 2: 519-522.?
[7] Karpathy A, Toderici G, Shetty S, et al. Large-scale video classification with convolutional neural networks[C]//Proceedings of the IEEE conference on Computer Vision and Pattern Recognition, 2014: 1725-1732.
[8] Lenc K, Vedaldi A. R-cnn minus r[EB/OL]. ArXiv preprint, 2015, arXiv:1506.06981.
[9] Girshick R. Fast r-cnn[C]//Proceedings of the IEEE international conference on computer vision, 2015: 1440-1448.
[10] Ren S, He K, Girshick R, et al. Faster R-CNN: towards real-time object detection with region proposal networks[J]. IEEE Transactions on Pattern Analysis & Machine Intelligence, 2017(6): 1137-1149.
[11] 吴晓凤, 张江鑫, 徐欣晨. 基于Faster R-CNN的手势识别算法[J]. 计算机辅助设计与图形学学报, 2018, 30(3): 468-476.
WU Xiaofeng, ZHANG Jiangxin, XU Xinchen. Hand Gesture Recognition Algorithm Based on Faster R-CNN[J]. Journal of Computer-Aided Design & Computer Graphics, 2018, 30(3): 468-476.
[12] Krizhevsky A, Sutskever I, Hinton G E. Imagenet classification with deep convolutional neural networks[C]//Advances in Neural Information Processing Systems, 2012: 1097-1105.
[13] Cire?an D, Meier U, Schmidhuber J. Multi-column deep neural networks for image classification[EB/OL]. Arxiv Preprint, 2012, arXiv:1202.2745.
[14] Szegedy C, Liu W, Jia Y, et al. Going deeper with convolutions[C]//Proceedings of the IEEE Conference On Computer Vision And Pattern Recognition, 2015: 1-9.
[15] Szegedy C, Vanhoucke V, Ioffe S, et al. Rethinking the inception architecture for computer vision[C]//Proceedings of the IEEE Conference On Computer Vision And Pattern Recognition, 2016: 2818-2826.
[16] Simonyan K, Zisserman A. Very deep convolutional networks for large-scale image recognition[EB/OL]. Arxiv Preprint, 2014, arXiv:1409.1556.
[17] Long J, Shelhamer E, Darrell T. Fully convolutional networks for semantic segmentation[C]//Proceedings of the IEEE conference on computer vision and pattern recognition, 2015: 3431-3440.
[18] LIN M, CHEN Q, YAN S. Network in network[EB/OL]. ArXiv preprint, 2013, arXiv: 1312.4400.
[19] Ioffe S, Szegedy C. Batch normalization: Accelerating deep network training by reducing internal covariate shift[EB/OL]. ArXiv preprint, 2015, arXiv: 1502.03167.
[20] Everingham M, Van Gool L, Williams C K I, et al. The pascal visual object classes (voc) challenge[J]. International Journal of Computer Vision, 2010, 88(2): 303-338.

备注/Memo

备注/Memo:
收稿日期:2019-11-19;修订日期:2020-04-24.
作者简介:于龙姣(1995-),女,硕士研究生,主要从事计算机视觉,利用深度学习的多目标检测算法研究。E-mail: yulongjiao@dlmu.edu.cn
基金项目:国家自然科学基金面上项目(61471079)。

更新日期/Last Update: 2020-07-16