Terahertz radiation is the part of the electromagnetic spectrum ranging from hundreds of MHz up to several THz. Recent advances in detector technology enable us to develop compact and affordable systems able to generate and detect THz radiation, thus filling the so-called “Terahertz gap”. Terahertz cameras with thousands of pixels arranged in one or two dimensions allow for real-time imaging applications. THz imaging techniques can be used to detect materials that are opaque to that radiation, such as metal or ceramic objects embedded in THz-transparent materials (e.g. paper, textiles, etc.). Due to its non-ionising nature, THz radiation does not require special shielding precautions compared to X-ray systems, and can be safely applied to living tissue. Terahertz imaging can be employed to detect forbidden objects in security checks with parcels scanners and full body scanners. This paper will describe our recent developments towards an innovative application of a sub-THz imaging system based on a commercially available detector. The setup is used to detect objects inside parcels and packages using a conveyor belt and a THz imaging system. An automatic detection system based on convolutional neural networks has been developed allowing for a real-time selection of targets against a set of images. The performance of the setup will be shown and newest results on the recognition capability will be presented.

Development of a Terahertz scanning apparatus with automatic object detection capabilities

Paoletti R.;Paolucci F.;
2020

Abstract

Terahertz radiation is the part of the electromagnetic spectrum ranging from hundreds of MHz up to several THz. Recent advances in detector technology enable us to develop compact and affordable systems able to generate and detect THz radiation, thus filling the so-called “Terahertz gap”. Terahertz cameras with thousands of pixels arranged in one or two dimensions allow for real-time imaging applications. THz imaging techniques can be used to detect materials that are opaque to that radiation, such as metal or ceramic objects embedded in THz-transparent materials (e.g. paper, textiles, etc.). Due to its non-ionising nature, THz radiation does not require special shielding precautions compared to X-ray systems, and can be safely applied to living tissue. Terahertz imaging can be employed to detect forbidden objects in security checks with parcels scanners and full body scanners. This paper will describe our recent developments towards an innovative application of a sub-THz imaging system based on a commercially available detector. The setup is used to detect objects inside parcels and packages using a conveyor belt and a THz imaging system. An automatic detection system based on convolutional neural networks has been developed allowing for a real-time selection of targets against a set of images. The performance of the setup will be shown and newest results on the recognition capability will be presented.
2020
Settore FIS/03 - Fisica della Materia
Millimetre Wave and Terahertz Sensors and Technology XIII 2020; Virtual, Online; United Kingdom; 21 September 2020 through 25 September 2020; Code 163800
gbr
2020
Volume 11541
SPIE-INT SOC OPTICAL ENGINEERING
9781510638952
CNN; Convolutional neural network; Security; Terahertz; YOLO
The Society of Photo-Optical Instrumentation Engineers (SPIE)
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11384/110504
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 3
  • ???jsp.display-item.citation.isi??? 1
  • OpenAlex ND
social impact