基于图像视野划分的公共场所人群计数模型

被引：8
|
作者：
袁健 [1 ]
王姗姗 [1 ]
罗英伟 [1 ]
机构：
[1] 上海理工大学光电信息与计算机工程学院
来源：
计算机应用研究 | 2021年 / 38卷 / 04期
关键词：
人数估计; 卷积神经网络; 图像视野划分; 轻量型;
D O I：
10.19734/j.issn.1001-3695.2020.02.0076
中图分类号：
TP391.41 []; TP183 [人工神经网络与计算];
学科分类号：
080203 ; 081104 ; 0812 ; 0835 ; 1405 ;
摘要：
为解决公共场所中人群分布不均以及目标尺度不一而影响人数估计的问题,提出了基于图像视野划分的公共场所人群计数模型。首先将图像场景划分为远近视野两个区域,对近视野区域,使用基于YOLO的网络进行行人检测并通过添加场景约束避免在远近视野区域内重复计数;对远视野区域,使用改进的MobileNets提取人群密度分布特征,并引入超分辨率重建模块提升人群密度图质量,最终通过计算两者之和得到整幅图像中的人群数量。在Shanghai Tech和Mall数据集上进行测试,结果表明该模型在准确性和鲁棒性上有显著的提高,实验证明模型切实可行。
引用
收藏
页码：1256 / 1260 +1280
页数：6
相关论文
共 20 条
  • [1] Real-time single image and video super-resolution using an efficient sub-pixel convolutional neural network . W.Shi,J.Caballero,F.Huszar,et al. IEEE Conference on Computer Vision and Pattern Recognition . 2016
  • [2] Depth Map Prediction from a Single Image Using a Multi-scale Deep Network . Eigen D,Puhrsch C,Fergus R. Proceedings of the 27th International Conference on Neural Information Processing Systems . 2014
  • [3] Learning to count objects in images . Lempitsky V,Zisserman A. Proceedings of the 23rd International Conference on Neural Information Processing Systems . 2010
  • [4] Slic superpixels . Achanta R,Shaji A,Smith K,Lucchi A,Fua P,Susstrunk S. EPFL-Report-149300 . 2010
  • [5] CSRNet:Dilated convolutional neural networks for understanding the highly congested scenes . Li Yuhong,Xiaofan Zhang,Deming Chen. 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition . 2018
  • [6] Cumulative attribute space for age and crowd density estimation . Chen Ke,Gong Shaogang,Xiang Tao et al. Proc of IEEE Conference on Computer Vision and Pattern Recognition . 2013
  • [7] Scale aggregation network for accurate and efficient crowd counting . Cao Xinkun,Wang Zhipeng,Zhao Yanyun et al. Proc of the15th European Conference on Computer Vision . 2018
  • [8] Single-image crowd counting via multi-column convolutional neural network . Zhang Yingying,Zhou Desen,Chen Siqi et al. Proc of IEEE Conference on Computer Vision and Pattern Recognition . 2016
  • [9] Cross-scene crowd counting via deep convolutional neural networks . Zhang Cong,Li Hongsheng,Wang Xiaogang et al. Proc of IEEE Conference on Computer Vision and Pattern Recognition . 2015
  • [10] Fully convolutional crowd counting on highly congested scenes . Marsden M,Mcguinness K,Little S et al. Proc of the 12th International Joint Conference on Computer Vision Imaging and Computer Graphics Theory and Applications . 2017
12