| 基于融合FPN和Faster R-CNN的行人检测算法 |
| |
| 引用本文： | 王飞,王林,张儒良,赵勇,王全红. 基于融合FPN和Faster R-CNN的行人检测算法[J]. 数据采集与处理, 2019, 34(3): 530-537 |
| |
| 作者姓名： | 王飞 王林 张儒良 赵勇 王全红 |
| |
| 作者单位： | 1.贵州民族大学人文科技学院，贵阳,550025;2.贵州民族大学数据科学与信息工程学院，贵阳,550025;3.北京大学深圳研究生院信息工程学院，深圳,518055 |
| |
| 基金项目： | 贵州省教育厅创新群体重大研究项目黔教合KY字[2018]018;深圳市科技计划JCYJ20160506172651253;贵州省研究生科研基金立项课题黔教研合KYJJ字[2016]04;贵州民族大学人文科技学院科研基金18rwjs016贵州省教育厅创新群体重大研究项目（黔教合KY字[2018]018）资助项目；深圳市科技计划（JCYJ20160506172651253）资助项目；贵州省研究生科研基金立项课题（黔教研合KYJJ字[2016]04）资助项目；贵州民族大学人文科技学院科研基金（18rwjs016）资助项目。 |
| |
| 摘 要： | 针对多尺度行人检测的问题，本文提出一种基于融合特征金字塔网络（Feature pyramid networks，FPN）和Faster R-CNN(Faster region convolutional neural network)的行人检测算法。首先，对FPN和区域建议网络（Region proposal networks，RPN）进行融合；然后，对FPN和Fast R-CNN进行融合；最后，在Caltech数据集、KITTI数据集和ETC数据集上分别对融合FPN和Faster R-CNN的行人检测算法进行训练和测试。该算法在Caltech数据集、KITTI数据集和ETC数据集上的mAP (mean Average Precision)分别达到69.72%， 69.76%和89.74%。与Faster R-CNN相比，该算法不仅提高了行人检测精度，而且在多尺度行人检测的问题上也获得了较为满意的检测效果。
|
| 关 键 词： | 特征金字塔网络 区域建议网络 Faster R-CNN 多尺度行人检测 |
| 收稿时间： | 2018-02-23 |
| 修稿时间： | 2019-04-19 |
Pedestrian Detection Algorithm Based on Fusion FPN and Faster R-CNN |
| |
| Wang Fei,Wang Lin,Zhang Ruliang,Zhao Yong,Wang Quanhong. Pedestrian Detection Algorithm Based on Fusion FPN and Faster R-CNN[J]. Journal of Data Acquisition & Processing, 2019, 34(3): 530-537 |
| |
| Authors: | Wang Fei Wang Lin Zhang Ruliang Zhao Yong Wang Quanhong |
| |
| Affiliation: | 1.College of Humanities & Sciences, Guizhou Minzu University, Guiyang, 550025, China;2.College of Data Science and Information Engineering, Guizhou Minzu University, Guiyang, 550025, China;3.School of Electronic and Computer Engineering,Shenzhen Graduate School Peking University, Shenzhen, 518055, China |
| |
| Abstract: | Aiming at the problem of multi-scale pedestrian detection, a pedestrian detection algorithm based on fusion feature pyramid networks (FPN) and faster R-CNN (Faster region convolutional neural network) is proposed. Firstly, FPN and region proposal networks (RPN) are fused. Secondly, FPN and Fast R-CNN are fused. Finally, the pedestrian detection algorithm with fusion FPN and Faster R-CNN is trained and tested on Caltech dataset, KITTI dataset, and ETC dataset, respectively. The mAP (mean Average Precision) of this algorithm reaches 69.72%, 69.76% and 89.74% on Caltech dataset, KITTI dataset, and ETC dataset, respectively. Compared with Faster R-CNN, this algorithm not only improves the pedestrian detection accuracy, but also obtains satisfactory detection effect on the problem of multi-scale pedestrian detection. |
| |
| Keywords: | feature pyramid networks region proposal networks Faster R-CNN (Faster region convolu-tional neural network) multi-scale pedestrian detection |
|
| 点击此处可从《数据采集与处理》浏览原始摘要信息 |
|
点击此处可从《数据采集与处理》下载全文 |
|
