首页 > 论文 > 激光与光电子学进展 > 55卷 > 12期(pp:121503--1)

改进的基于卷积神经网络的人数估计方法

Improved Method for Estimating Number of People Based on Convolution Neural Network

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

摘要

估算监控场景中的人数是安防监控的重要任务之一, 当人群密集、行人之间存在遮挡时, 人数估计较困难。因此, 针对密集场景下的人数估计问题, 提出了一种改进的基于卷积神经网络的人数估计方法。为了改善摄像透视畸变带来的影响, 分别利用深层网络和浅层网络提取人群特征, 深层和浅层网络分别设计了不同核大小的卷积层, 并将提取到的特征通过一个具备多尺度提取能力的结构进行融合。实验结果表明, 改进后的网络模型所获取的人群密度图更加贴近原场景信息, 人数估计结果也更加精确。

Abstract

Estimating the number of people in the surveillance scene is one of the important tasks of security monitoring. However it is difficult to estimate the number when the crowd is with clutter and severe occlusion. An improved crowd counting method based on the convolution neural network is proposed as for the number estimation under dense scenes. In order to reduce the effect of camera perspective distortion, the deep network and shallow network are used to extract the crowd characteristics, respectively. The convolution layers with different kernel sizes are also designed. Moreover, the extracted features are fused through a special structure with multi-scale extraction capability. The experimental results show that the crowd density map obtained by the improved network model is closer to the original scene information and the obtained prediction results are more precise.

中国激光微信矩阵
补充资料

中图分类号:TP391

DOI:10.3788/LOP55.121503

所属栏目:机器视觉

基金项目:国家自然科学基金民航联合研究基金重点项目(U1533203)、中央高校基本科研业务费项目中国民航大学专项资助(3122017005, 3122018C004)

收稿日期:2018-05-14

修改稿日期:2018-06-14

网络出版日期:2018-07-05

作者单位    点击查看

张红颖:中国民航大学电子信息与自动化学院, 天津 300300
王赛男:中国民航大学电子信息与自动化学院, 天津 300300
胡文博:中国民航大学电子信息与自动化学院, 天津 300300

联系人作者:张红颖(carole_zhang0716@163.com)

【1】Yin Z Q, Gu G H, Chen Q, et al. Application of three-dimensional depth image in automatic statistic passenger flow system[J]. Chinese Journal of Lasers, 2014, 41(6): 0609003.
尹章芹, 顾国华, 陈钱, 等. 三维深度图像在自动客流计数系统中的应用[J]. 中国激光, 2014, 41(6): 0609003.

【2】Ryan D, Denman S, Sridharan S, et al. An evaluation of crowd counting methods, features and regression models[J]. Computer Vision and Image Understanding, 2015, 130: 1-17.

【3】Sindagi V A, Patel V M. A survey of recent advances in CNN-based single image crowd counting and density estimation[J]. Pattern Recognition Letters, 2018, 107: 3-16.

【4】Chen K, Loy C C, Gong S, et al. Feature mining for localised crowd counting[C]∥Proceedings of the 23rd British Machine Vision Conference, 2012, 1(2): 3.

【5】Chan A B, Vasconcelos N. Counting people with low-level features and Bayesian regression[J]. IEEE Transactions on Image Processing, 2012, 21(4): 2160-2177.

【6】Liang R H, Liu X D, Ma X Y, et al. High-density crowd counting method based on SURF feature[J]. Journal of Computer-Aided Design & Computer Graphics, 2012, 24(12): 1559-1567.
梁荣华, 刘向东, 马祥音, 等. 基于SURF的高密度人群计数方法[J]. 计算机辅助设计与图形学学报, 2012, 24(12): 1559-1567.

【7】Zhang Y Y, Zhou D S, Chen S Q, et al. Single-image crowd counting via multi-column convolutional neural network[C]∥Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, 2016: 589-597.

【8】Zhang C, Li H S, Wang X G, et al. Cross-scene crowd counting via deep convolutional neural networks[C]∥Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, 2015: 833-841.

【9】Boominathan L, Kruthiventi S S S, Babu R V. Crowdnet: a deep convolutional network for dense crowd counting[C]∥Proceedings of the 2016 ACM on Multimedia Conference, 2016: 640-644.

【10】Chen L C, Papandreou G, Kokkinos I, et al. Deeplab semantic image segmentation with deep convolutional nets, atrous convolution, and fully connected CRFs[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2018, 40(4): 834-848.

【11】Chu J L, Krzyak A. Analysis of feature maps selection in supervised learning using convolutional neural networks[C]∥Proceedings of Canadian Conference on Artificial Intelligence, 2014: 59-70.

【12】Krizhevsky A, Sutskever I, Hinton G E. Imagenet classification with deep convolutional neural networks[C]∥Proceedings of the 25th International Conference on Neural Information Processing Systems, 2012: 1097-1105.

【13】Simonyan K, Zisserman A. Very deep convolutional networks for large-scale image recognition[J]. arXiv preprint arXiv: 1409.1556, 2014.

【14】Idrees H, Saleemi I, Seibert C, et al. Multi-source multi-scale counting in extremely dense crowd images[C]∥Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, 2013: 2547-2554.

【15】Hu Y C, Chang H, Nian F D, et al. Dense crowd counting from still images with convolutional neural networks[J]. Journal of Visual Communication and Image Representation, 2016, 38: 530-539.

【16】Lempitsky V, Zisserman A. Learning to count objects in images[C]∥Proceedings of International Conference on Neural Information Processing Systems, 2010: 1324-1332.

【17】Jones M J, Snow D. Pedestrian detection using boosted features over many frames[C]∥Proceedings of 19th International Conference on Pattern Recognition, 2008: 1-4.

【18】Felzenszwalb P, McAllester D, Ramanan D. A discriminatively trained, multiscale, deformable part model[C]∥Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, 2008: 1-8.

引用该论文

Zhang Hongying,Wang Sainan,Hu Wenbo. Improved Method for Estimating Number of People Based on Convolution Neural Network[J]. Laser & Optoelectronics Progress, 2018, 55(12): 121503

张红颖,王赛男,胡文博. 改进的基于卷积神经网络的人数估计方法[J]. 激光与光电子学进展, 2018, 55(12): 121503

被引情况

【1】吴沛佶,梅雪,何毅,袁申强. 基于深度网络模型的视频序列中异常行为的检测方法. 激光与光电子学进展, 2019, 56(13): 131101--1

【2】马晓云,朱丹,金晨,佟新鑫. 基于改进Faster R-CNN的子弹外观缺陷检测. 激光与光电子学进展, 2019, 56(15): 151202--1

【3】吉祥凌,吴军,易见兵,张晓光. 基于深度学习的管制物品自动检测算法研究. 激光与光电子学进展, 2019, 56(18): 180402--1

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