Terahertz video analysis for hidden object detection using deep learning techniques

Abstract

This paper is based on automatically detecting hazardous and non-dangerous objects from terahertz images. First, we trained a neural network to automatically analyze dangerous and non-dangerous items, which can be used for experiment with terahertz images generated by a prototype terahertz video system. The system comprises a terahertz video database of people hiding dangerous and non-dangerous objects under their clothing. Secondly, visual geometry group-19 (VGG-19) is trained on dangerous and non-dangerous objects from our database. After training, the accuracy received was 99.6% for safe items and 85.85% for dangerous items. We tested the network with various categories of objects not included in the training set and found that most were correctly identified as dangerous and nondangerous items. Also, we have identified some of the critical issues that need to be addressed to make this technology more accessible and widely used. Our work can pave the way for future research in this field and help to address the challenges associated with terahertz imaging technologies. The paper describes some preliminary terahertz video surveillance experiments necessary for developing a natural terahertz video surveillance system.

Keywords

A

References

[1] Yang M X, Zhao Y M, Liu H X, et al. . A novel weakly supervised learning framework for hazardous item localization in terahertz security imaging. Infrared physics & technology. 2024, 140: 1350-4495[J]

[2] Cheng R, Lucyszyn S. Few-shot concealed object detection in sub-THz security images using improved pseudoannotations. Scientific reports. 2024, 14(1): 3150 J]

[3] Chandel S, Bhatnagar G, Kowalski M. Saliency and superpixel improved detection and segmentation of concealed objects for passive terahertz images. Optical engineering. 2023, 62(2): 023101 J]

[4] SARDAR S, ASSI S, ZOLKIFLY I A, et al. Application of deep learning algorithms to terahertz images for detection of concealed objects[J]. Lecture notes on data engineering and communications technologies, 2024: 279–289.

[5] Ge Z, Zhang Y, Jiang Y, et al. . Lightweight YOLOv7 algorithm for multi-object recognition on contrabands in terahertz images. Applied sciences. 2024, 14(4): 1398 J]

[6] Danso S A, Shang L, Hu D, et al. . Hidden dangerous object recognition in terahertz images using deep learning methods. Applied sciences. 2022, 12(15): 7354 J]

[7] CHAKRABORTY S, PALODHI K. Introduction to terahertz imaging applications[J]. Optical to terahertz engineering, 2023, 1–15.

[8] Bharadwaj A N, Kashyap A M, Jayachandran R, et al. . A survey on terahertz devices – a cutting edge technology. Proceedings of the International Conference on Recent Trends in Electronics and Communication. 2023, New York, IEEE: 1-6[C]

[9] Bagchi P, Sushkova O S, Morozov A A, et al. . TCDLN: terahertz colormap deep learning network for hidden object detection. Electrical systems. 2024, 20(3): 2567-2583[J]

[10] Terahertz security body scanner. https://TeraSense.com/products/body-scanner/.

[11] Astrohn technology. https://astrohn.eu/.

[12] Thruvision. https://thruvision.com/.

[13] Development of the multi-agent logic programming approach to human behavior analysis in a multi-channel video surveillance. https://www.fullvision.ru/.

[14] Chen Z, Tian R, Xiong D, et al. . Multi-dimensional information fusion you only look once network for suspicious object detection in millimeter wave images. Electronics. 2024, 13(4): 773 J]

[15] Cheng L, Zhang W, Li C, et al. . HE-DETR-DC5 method to rapidly detect objects on passive THz images in human safety inspection. Electronics letters. 2023, 59(2): e12702 J]

[16] Cheng L, Ji Y, Li C, et al. . Improved SSD network for fast concealed object detection and recognition in passive terahertz security images. Scientific reports. 2022, 12(1): 12082 J]

[17] Hocine M, Hamid Z, Mohamed L. Concealed objects detection and recognition in active transmission sub-terahertz imaging scanner for automatic security screening. Optical engineering. 2024, 63(3): 033102[J]

[18] Sen P, Hall J, Polese M, et al. . Terahertz communications can work in rain and snow: impact of adverse weather conditions on channels at 140 GHz. Proceedings of the 6th ACM Workshop on Millimeter-Wave and Terahertz Networks and Sensing Systems, 2022. 202213-18[C]

RIGHTS & PERMISSIONS

Tianjin University of Technology