Occluded Prohibited Items Detection: An X-ray Security Inspection Benchmark and De-occlusion Attention Module

We present an X-ray material classifier region-based convolutional neural network (XMC R-CNN) model for detecting the typical guns and the typical knives in X-ray baggage images. The XMC R-CNN model is used to solve the problem of contraband detection in overlapped X-ray baggage images by the X-ray material classifier algorithm and the organic stripping and inorganic stripping algorithm, and better detection rate and the miss rate are achieved. The detection rates of guns and knives are 96.5% and 95.8%, and the miss rates of guns and knives are 2.2% and 4.2%. The contraband detection technology based on the XMC R-CNN model is applied to X-ray baggage images of security inspection. According to user needs, the safe X-ray baggage images can be automatically filtered in some specific fields, which reduces the number of X-ray baggage images that security inspectors need to screen. The efficiency of security inspection is improved, and the labor intensity of security inspection is reduced. In addition, the security inspector can screen X-ray baggage images according to the boxes of automatic detection, which can improve the effect of security inspection.

Download Full-text

Detail Restoration and Tone Mapping Networks for X-Ray Security Inspection

IEEE Access ◽

10.1109/access.2020.3035086 ◽

2020 ◽

Vol 8 ◽

pp. 197473-197483

Author(s):

Hyo-Young Kim ◽

Seung Park ◽

Yong-Goo Shin ◽

Seung-Won Jung ◽

Sung-Jea Ko

Keyword(s):

Tone Mapping ◽

X Ray ◽

Security Inspection

Download Full-text

Temporal Effects in a Security Inspection Task: Breakdown of Performance Components

Proceedings of the Human Factors and Ergonomics Society Annual Meeting ◽

10.1177/154193120705100209 ◽

2007 ◽

Vol 51 (2) ◽

pp. 93-97 ◽

Cited By ~ 6

Author(s):

K. M. Ghylin ◽

C. G. Drury ◽

R Batta ◽

L. Lin

Keyword(s):

False Alarm ◽

Stopping Time ◽

Probability Of Detection ◽

Vigilance Decrement ◽

X Ray ◽

Temporal Effects ◽

Probability Of False Alarm ◽

Active Search ◽

Security Inspection ◽

Inspection Task

Data from certified screeners performing an x-ray inspection task for 4 hours, or 1000 images, were analyzed to identify the nature of the vigilance decrement. The expected vigilance decrement was found, with performance measured by probability of detection (PoD) and probability of false alarm [P(FA)] decreasing from hour 1 to hour 4. Correlations between PoD and P(FA) indicate that sensitivity between hours remained the same, however a shift in criterion (Beta) occurred. Significant decreases in both detection and stopping time were found from the first hour to the second, third, and fourth hour. Evidence of changes in the search component of the time per item was found to account for part of the vigilance decrement. As the task continued, participants spent less time actively searching the image, as opposed to other activities. Evidence is provided for truncation of active search as security inspection continues.

Download Full-text

Deep Feature Fusion Based Dual Branch Network for X-ray Security Inspection Image Classification

Applied Sciences ◽

10.3390/app11167485 ◽

2021 ◽

Vol 11 (16) ◽

pp. 7485

Author(s):

Yingda Xu ◽

Jianming Wei

Keyword(s):

Computer Security ◽

Network Architecture ◽

Feature Fusion ◽

Class Imbalance ◽

Classification Performance ◽

Fusion Model ◽

X Ray ◽

Deep Feature ◽

Scanned Images ◽

Security Inspection

Automatic computer security inspection of X-ray scanned images has an irresistible trend in modern life. Aiming to address the inconvenience of recognizing small-sized prohibited item objects, and the potential class imbalance within multi-label object classification of X-ray scanned images, this paper proposes a deep feature fusion model-based dual branch network architecture. Firstly, deep feature fusion is a method to fuse features extracted from several model layers. Specifically, it operates these features by upsampling and dimension reduction to match identical sizes, then fuses them by element-wise sum. In addition, this paper introduces focal loss to handle class imbalance. For balancing importance on samples of minority and majority class, it assigns weights to class predictions. Additionally, for distinguishing difficult samples from easy samples, it introduces modulating factor. Dual branch network adopts the two components above and integrates them in final loss calculation through the weighted sum. Experimental results illustrate that the proposed method outperforms baseline and state-of-art by a large margin on various positive/negative ratios of datasets. These demonstrate the competitivity of the proposed method in classification performance and its potential application under actual circumstances.

Download Full-text

GAN-based data augmentation of prohibited item X-ray images in security inspection

Optoelectronics Letters ◽

10.1007/s11801-020-9116-z ◽

2020 ◽

Vol 16 (3) ◽

pp. 225-229

Author(s):

Yue Zhu ◽

Hai-gang Zhang ◽

Jiu-yuan An ◽

Jin-feng Yang

Keyword(s):

Data Augmentation ◽

X Ray ◽

Security Inspection

Download Full-text

X-Ray Security Inspection Image Detection Based on a Multi-scale Feature Fusion Network

Advances in Natural Computation, Fuzzy Systems and Knowledge Discovery - Lecture Notes on Data Engineering and Communications Technologies ◽

10.1007/978-3-030-89698-0_83 ◽

2022 ◽

pp. 814-821

Author(s):

Zhuang Qiao ◽

Hong Zhang

Keyword(s):

Feature Fusion ◽

Image Detection ◽

X Ray ◽

Scale Feature ◽

Multi Scale ◽

Security Inspection

Download Full-text

X-ray Security Inspection Image Detection Algorithm Based on Improved YOLOv4

10.1109/ecice52819.2021.9645636 ◽

2021 ◽

Author(s):

Cheng Zhou ◽

Hui Xu ◽

Bicai Yi ◽

Weichao Yu ◽

Chenwei Zhao

Keyword(s):

Detection Algorithm ◽

Image Detection ◽

X Ray ◽

Security Inspection

Download Full-text

Program Peningkatan Keamanan Di Bandara Kualanamu Medan Apabila Beroperasi

WARTA ARDHIA ◽

10.25104/wa.v38i1.215.97-105 ◽

2012 ◽

Vol 38 (1) ◽

pp. 97-105

Author(s):

Ismail Nadjamuddin

Keyword(s):

Detection System ◽

Aviation Security ◽

Closed Circuit ◽

Airport Security ◽

Civil Aviation ◽

Metal Detector ◽

Closed Circuit Television ◽

X Ray ◽

Security Officer ◽

Security Inspection

Kualanamu Medan Airport Construction is an alternative to overcome the operational limitations and the land that was experienced Polonia Airport. Airport security enhancement program Kualanamu Medan when operating include airport security aspects of the provision of facilities consisting of: X-Ray facility 12 units, Walk Throught Metal Detector (WTMD) 14 units, Hand Held Metal Detector (HHMD) 24 units, Closed Circuit Television (CCTV , Explosive Detection System 2 units, 2 units Liquit Scan Detecto,, Detector Nubikara 2 units (Nuclear, Biological, Chemical, Radio Active), Body Scane 1 unit, Body Inspector, Airport For Perimeter Surveillance, Security Inspection Car and Motorcycle 4. The security officer in Medan Airport Kualanamu totaling 204 personnel consisting of 124 personnel from the PT. Angkasa PuraII, 40 BKO-military personnel who assisted and 40 staff personnel outsoursing. The system and airport security procedures will refer to the Regulation of the Minister of Transportation No. 9 of 2010 about the National Aviation Security Programme and ICAO in Annex17 on Security and Document-8973 on the Security Manual for Safeguarding Civil Aviation Against Acts of Unlawful Interference, that the safety and security systems at airports should be the maximum, using equipment and adequate procedures that ensure safety and smooth flight. Pembangunan Bandara Kualanamu Medan merupakan alternatif untuk mengatasi keterbatasan operasional dan tanah yang dialami Bandara Polonia. Program peningkatan keamanan bandara Kualanamu Medan saat dioperasikan meliputi aspek keamanan bandara, penyediaan fasilitas yang terdiri dari: 12unit fasilitas X-Ray, 14 unit Walking Through Mental Detector (WTMD), 24 unit Hand Held Metal Detector (HHMD), 2 unit Closed Circuit Television (CCTV)dan sistem Deteksi peledak, 2 unit Liquit Pindai Detecto, 2 unit Detector Nubikara (Nuklir, Biologi, Kimia, Radio Aktif), Badan Scane 1 unit, Bandara Untuk Surveillance Perimeter, 4 Mobil Keamanan Inspeksi dan Sepeda Motor petugas keamanan di Medan Bandara Kualanamu sebesar 204 personel yang terdiri dari 124 personil dari PT Angkasa Pura II, 40 BKO-personil militer yang dibantu dan 40 personil staf outsoursing. Sistem dan prosedur keamanan bandara akan mengacu pada Peraturan Menteri Perhubungan Nomor 9 tahun 2010 tentang National Aviation Security Program dan ICAO dalam Annex-17 tentang Keamanan dan manual dokumen-8973 tentang Keamanan Penerbangan Sipil upaya pelanggaran hukum, bahwa keamanan dan sistem keamanan di bandara harus maksimal , menggunakan peralatan dan prosedur yang memadai yang menjamin keamanan dan kelancaran penerbangan.

Download Full-text