北京航空航天大学学报 ›› 2019, Vol. 45 ›› Issue (10): 2011-2025.doi: 10.13700/j.bh.1001-5965.2019.0019

• 论文 • 上一篇    下一篇

PMMWI与VI优势互补的人体隐蔽违禁物检测与定位

赵国, 秦世引   

  1. 北京航空航天大学 自动化科学与电气工程学院, 北京 100083
  • 收稿日期:2019-01-17 出版日期:2019-10-20 发布日期:2019-10-31
  • 通讯作者: 秦世引 E-mail:qsy@buaa.edu.cn
  • 作者简介:赵国 男,博士研究生。主要研究方向:深度学习、图像语义分割、目标的检测与识别;秦世引 男,博士,教授,博士生导师。主要研究方向:图像处理、模式识别、智能优化控制等。
  • 基金资助:
    国家自然科学基金(61731001)

Detection and localization of concealed forbidden objects on human body based on complementary advantages of PMMWI and VI

ZHAO Guo, QIN Shiyin   

  1. School of Automation Science and Electrical Engineering, Beihang University, Beijing 100083, China
  • Received:2019-01-17 Online:2019-10-20 Published:2019-10-31
  • Supported by:
    National Natural Science Foundation of China (61731001)

摘要: 根据公共场所人体安检的性能要求和技术需求,将被动毫米波成像(PMMWI)的可透视成像性能优势与可见光成像(VI)的细节高分辨性能优势相结合,提出一种基于PMMWI与VI优势互补的人体隐蔽违禁物检测与定位算法。首先,提出一种基于低层特征融合的改进U-Net以增强深度神经网络(DNN)对PMMWI中弱小目标轮廓的敏感度,提高PMMWI中人体轮廓和隐蔽违禁物的分割精度,并同时实现VI中人体轮廓的像素级分割;然后,在PMMWI和VI中的人体轮廓分割基础上,通过基于人体轮廓的尺度变换与滑动适配实现PMMWI人体轮廓和VI人体轮廓的良好配准,根据配准结果实现单帧图像中人体隐蔽违禁物的高效检测;最后,通过序列图像检测结果的对比融合与优化决策给出隐蔽违禁物的定位结果。一系列综合实验与对比分析结果,验证了提出的人体隐蔽违禁物检测与定位算法的性能优势。

关键词: 毫米波安检, 被动毫米波成像(PMMWI), 人体轮廓分割, 深度学习, 深度神经网络(DNN), 隐蔽违禁物检测与定位

Abstract: According to the performance requirements and technology demands of human security check in public place, combining the performance advantages of perspective imaging of passive millimeter wave imaging (PMMWI) and high discriminability in image details of visible imaging (VI), an approach for detection and localization of concealed forbidden objects on human body is presented in this paper based on the complementary advantages of PMMWI and VI. Firstly, an improved U-Net based on feature fusion in low layers is presented to enhance the sensitivity of deep neural networks (DNN) to the contour of dim small targets, and improve the accuracy of segmentation of human contours and concealed forbidden objects. Meanwhile, the pixel-level segmentation of human contours in VI is also implemented. Then, the contours of human body in PMMWI are registered with some corresponding ones in VI by scale transform and sliding fit, so the concealed forbidden objects on human body can be detected from a single frame with a high precision. Finally, the concealed forbidden objects are localized by contrasting fusion and optimizing decision according to detection results with sequence images. A series of comprehensive experiments and comparative analysis results validate the good performance of the proposed detection and localization algorithm of concealed forbidden objects on human body towards security check of public places.

Key words: millimeter wave security check, passive millimeter wave imaging (PMMWI), contour segmentation of human body, deep learning, deep neural network (DNN), detection and localization of concealed forbidden objects

中图分类号: 


版权所有 © 《北京航空航天大学学报》编辑部
通讯地址:北京市海淀区学院路37号 北京航空航天大学学报编辑部 邮编:100191 E-mail:jbuaa@buaa.edu.cn
本系统由北京玛格泰克科技发展有限公司设计开发