Deep CNNs for Object Detection Using Passive Millimeter Sensors

被引:39
作者
Lopez-Tapia, Santiago [1 ]
Molina, Rafael [1 ]
Perez de la Blanca, Nicolas [1 ]
机构
[1] Univ Granada, Dept Comp Sci & Artificial Intelligence, Granada 18010, Spain
关键词
Classification; deep learning; millimeter wave imaging; object detection; security; CONCEALED OBJECTS; NEURAL-NETWORKS;
D O I
10.1109/TCSVT.2017.2774927
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
Passive millimeter wave images (PMMWIs) can be used to detect and localize objects concealed under clothing. Unfortunately, the quality of the acquired images and the unknown position, shape, and size of the hidden objects render these tasks challenging. In this paper, we discuss a deep learning approach to this detection/localization problem. The effect of the nonstationary acquisition noise on different architectures is analyzed and discussed. A comparison with shallow architectures is also presented. The achieved detection accuracy defines a new state of the art in object detection on PMMWIs. The low computational training and testing costs of the solution allow its use in real-time applications.
引用
收藏
页码:2580 / 2589
页数:10
相关论文
共 42 条
[1]   Radiation exposure and privacy concerns surrounding full-body scanners in airports [J].
Accardo, Julie ;
Chaudhry, M. Ahmad .
JOURNAL OF RADIATION RESEARCH AND APPLIED SCIENCES, 2014, 7 (02) :198-200
[2]   Multispectral mm-wave imaging:: materials and images [J].
Alexander, Naomi E. ;
Andres, Carlos Callejero ;
Gonzalo, Ramon .
PASSIVE MILLIMETER-WAVE IMAGING TECHNOLOGY XI, 2008, 6948
[3]   Compressive sensing super resolution from multiple observations with application to passive millimeter wave images [J].
AlSaafin, Wael ;
Villena, Salvador ;
Vega, Miguel ;
Molina, Rafael ;
Katsaggelos, Aggelos K. .
DIGITAL SIGNAL PROCESSING, 2016, 50 :180-190
[4]  
[Anonymous], 2015, ARXIV 1502 03167
[5]  
[Anonymous], 2015, TECH REP
[6]  
[Anonymous], 2016, ADAPT COMPUT MACH LE
[7]  
[Anonymous], IEEE T PATTERN ANAL
[8]  
[Anonymous], cloud based on convolut ional neural network
[9]  
[Anonymous], 2015, P INT C LEARN REPR, DOI DOI 10.48550/ARXIV.1511.07289
[10]  
[Anonymous], 2013, PMLR