A novel enhanced region proposal network and modified loss function: threat object detection in secure screening using d
- PDF / 3,504,811 Bytes
- 30 Pages / 439.37 x 666.142 pts Page_size
- 61 Downloads / 150 Views
A novel enhanced region proposal network and modified loss function: threat object detection in secure screening using deep learning Priscilla Steno1 · Abeer Alsadoon1 · P. W. C. Prasad1 · Thair Al‑Dala’in1 · Omar Hisham Alsadoon2
© Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract Detection of threat objects concealed in passenger clothing and baggage poses a challenge to aviation security. At present, the detection technology is capable of detecting the presence of threats from the scanned images yet requires the involvement of human in determining what type of threat and where it is located. Deep learning-based object detection technique has not been successfully implemented to detect threats in the security screening processes. This research aims to improve the accuracy and the processing time of threat detection in security screening. Enhanced faster region-based convolutional neural network (faster R-CNN) with improved region proposals is used for better threat localization. The proposed system consists of an improved region proposal network that outputs object’s region proposals with an object score to the detector module to accurately locate the threat in the human body. Furthermore, this system uses a modified loss function that strengthens the classification loss. Results obtained by the proposed model show a 15% improvement in object localization. Therefore, the enhanced faster R-CNN achieves an overall detection accuracy of 0.27 in terms of average precision and reduces the processing time by 0.19 s. The results obtained by the enhanced faster R-CNN for detection accuracy are superior to the state-of-the-art system. Also, this model focuses on localizing the threat and identifying its type, which makes the model suitable for threat detection security screening. Besides, the research also addresses the time consumption issue in detecting the threat object. Keywords Threat detection · Region proposal network · Deep learning · Faster R-CNN · Cross-entropy loss · Airport security
* Abeer Alsadoon [email protected] 1
School of Computing and Mathematics, Charles Sturt University, Sydney, Australia
2
Department of Islamic Sciences, Al Iraqia University, Baghdad, Iraq
13
Vol.:(0123456789)
P. Steno et al.
1 Introduction Transport Security Administration (TSA) operates an Advanced Imaging Technology (AIT) unit with a target recognition software for passenger screening [1]. This creates a generic outline of the human body on which a yellow box appears if the scanner detects an object. Whenever a potential threat is identified, a TSA officer is required to engage in a secondary, manual security screening process. The manual process creates significant bottlenecks at the checkpoints; further threat prediction algorithm is inaccurate in predicting the region where the threat is located [2]. According to Guimaraes and Tofighi [2], the Department of Homeland Security in the USA reports higher false alarm rates resulting from their passenger-screening algorithm used at
Data Loading...
