Research on person re-identification algorithm based on soft mask foreground segmentation and multi information fusion re-ranking

doi:10.13741/j.cnki.11-1879/o4.2023.06.010

您当前的位置：

首页 >

文章列表页 >

Research on person re-identification algorithm based on soft mask foreground segmentation and multi information fusion re-ranking

更新时间：2025-09-01

- Research on person re-identification algorithm based on soft mask foreground segmentation and multi information fusion re-ranking
- Vol. 49, Issue 6, Pages: 756-762(2023)
- 作者机构：
  
  1:北京理工大学集成电路与电子学院
- 作者简介：
- 基金信息：
- DOI：10.13741/j.cnki.11-1879/o4.2023.06.010
  CLC： TP391.41
- Published：2023
- 稿件说明：
移动端阅览
Research on person re-identification algorithm based on soft mask foreground segmentation and multi information fusion re-ranking[J]. 2023, 49(6): 756-762.
DOI：

Research on person re-identification algorithm based on soft mask foreground segmentation and multi information fusion re-ranking[J]. 2023, 49(6): 756-762. DOI： 10.13741/j.cnki.11-1879/o4.2023.06.010.

摘要

为了提升行人重识别算法的检索准确率，提出了基于软掩膜前景分割和多信息融合重排序的行人重识别算法。通过基于U-Net的软掩膜前景分割算法去除图像背景信息，同时减缓图像分割边界的突变，以保留图像中行人的关键信息；通过基于孪生深度网络的多信息融合重排序算法融合行人图像的多种信息，提升检索结果质量。实验结果表明，提出的两种方法是对行人重识别算法流程的有效补充，可作为提升准确率的有效方法应用于多数现有行人重识别算法中。

Abstract

关键词

Keywords

references

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Weak supervised abnormal behavior detection using improved YOLOv3 under video surveillance

Abnormal crowd behavior detection using improved C3D-RF under Video Surveillance

Design of surveillance system for real-time tracking of moving object based on DM642

Design of intelligent video monitoring system based on USB 2.0

Person re-identification method based on Visible-Infrared modal bidirectional feature generation

Related Author

赵雪章

丁犇

席运江

张娓娓

陈绥阳

陈锐

谢永昌

许少凡

Related Institution

1:佛山职业技术学院电子信息学院

2:华南理工大学经济管理学院

1:西安思源学院电子信息工程学院

2:西安交通大学理学院

3:郑州轻工业大学软件学院

⁰