Computed Tomography Advanced Technology vs. Multiview/Dual View X-ray and Other Technologies: Enhancing Checkpoint Security

Terrorists continue to target commercial aviation. Over the past two years, there have been several direct attacks against aircraft, including the 2015 downing of Metrojet Flight 9268 in Egypt and the 2016 attempted downing of Daallo Airlines Flight 159 in Somalia, along with armed landside attacks in Brussels and Istanbul. Intelligence reports also indicate that terrorist groups have improved their ability to hide Improvised Explosive Devices in ways that can fool today’s X-ray scanners by, for example, concealing explosive devices within personal electronic devices (PEDs) larger than a mobile phone. Recently, based on a lack of confidence in existing measures for cabin baggage screening, the authorities have banned PEDs in cabin baggage on routes from ten major Middle Eastern airports to the United States and six departure points to the United Kingdom, creating anger and confusion for travellers. Joe Paresi considers the capability of available technology to resolve regulators’ concerns and keep airline patrons happy.

Thin sheet explosives hidden in electronics
Fig. 1: Thin sheet explosives hidden in electronics can easily be identified in PEDs with CT

As threats continue to evolve, new technology is desperately needed. Existing conventional and AT X-ray scanners used at many large airports face challenges in addressing these evolving threats as they have limited views and inadequate material discrimination capabilities. Also, X-ray scanners use a crude measurement of the Effective Atomic Number (Zeff) based on the ratio of high energy to low energy attenuation. Since X-rays pass through every object in its path, the resulting atomic number calculation is the average of all the objects it passed through; individual objects cannot be isolated by projection X-ray systems. This limits the probability of detection and leads to excessive false alarms. That is why these systems are normally used in an operator-assist rather than an automatic detection mode. In an attempt to help AT operators identify explosives hidden in electronics, passengers place their electronics in a separate tray to simplify the image so that a second side view can better help security officers identify anomalies, such as thin sheet explosives hidden inside a laptop computer. However, since only two views are present and other IED components (timers, detonators) can easily be hidden amongst the typical electronic circuitry inside such electronics, this approach has not been sufficiently effective.

Next Generation Technologies and Trade-offs

Multiview AT technology available from suppliers today use several X-ray sources and detector arrays to image a bag from different orientations. For carry-on baggage applications, a maximum of only four views are used to date due to size and cost constraints. Systems viewing bags from these four angles can only roughly estimate the size and material properties of each object, or suspected explosive, in the bag. System designers use this to determine if an object is large enough to be of concern (as too small an explosive charge is not considered a threat). Because these measurements are only approximations, these systems usually have an Operational False Alarm Rate (Ops FAR) as high as 30-40% in order to meet detection requirements.

“…if one took four images from four different angles [of a person’s head], then only an imprecise estimate of the size and shape of the head and facial features could be obtained…”

By way of analogy, consider taking photographs of an object from four different angles. An easy example would be that of a person’s head. If one took four images from four different angles, then only an imprecise estimate of the size and shape of the head and facial features could be obtained, even if the views were 90 degrees apart. Multiview X-ray scanners take four independent images, but at much smaller angles than 90°, as the size of the tunnel and the conveyor system constrain the location of the X-ray and opposing detector arrays. Two views are usually placed vertically through the bag and horizontally across the belt, supplemented by these two additional angled views. These two added views are limited by the physical width of the scanner to approximately 17° to 20°. It is these imperfect measurements of objects in a bag that result in high false alarm rates, which are hard to resolve with the cluttered, superimposed 2D images, especially if those threats are hidden behind items such as electronics. Thus, these systems are difficult to use operationally as it then takes a significant amount of time to resolve each alarm.

While it is possible for an AT system to pass less challenging government tests within the specified false alarm limit, it is not feasible to expect any of these systems to meet the current ECAC C-3/4 that includes liquids and advanced threats. All suppliers understand these limitations and ALL have their own strategies to address them. Some are trying to convert their existing dual-energy X-ray detectors to multi-energy, in order to provide an energy signature based on the captured photons at multiple energy levels. These detectors replace the photodiodes with Cadmium Telluride (CdTe) detectors, and the required associated electronics. CdTe detectors measure the photon energy levels at small intervals across the X-ray energy band (i.e. from 0 Kilovolts to 160 Kilovolts) by measuring the quantity of photons at each energy level captured by the detector, and comparing it to the energy level of explosive material measured in laboratory conditions. The benefit of these higher cost detectors to the system is so far unproven, and does not overcome the fundamental challenge of limited views, clutter and superimposition from other items within the bag.

“…a simple way to think about a CT scan of a bag is to imagine cutting a loaf of bread into slices and then cutting each slice into small cubes that can later be reconnected together to re-form a complete loaf…”

Another approach being developed uses X-ray diffraction (XRD), which measures the scatter pattern of X-rays based on the crystalline structure of the explosive material. Unless costly, high intensity X-ray tubes are used, X-ray diffraction is slow, requires several seconds of measurement data and its ability to measure thin sheet explosives, where the X-rays will pass through with limited scatter, is unproven. Coherent scatter also does not work accurately on continuous liquids. That is why most of the major security manufacturers are investing in the further development of Computed Tomography (CT) systems for the checkpoint security market. Additionally, it is the only proven method of obtaining measurements of the entire content of a bag accurately, quickly and cost-effectively, whilst also providing a full three-dimensional image for operator review.

Computed Tomography

Computed Tomography, or CT, is an imaging approach that collects significantly more views of an object within a bag. Compared to the four or less projection images/views for AT technology, CT technology takes as many as 1440 images. Reconstruction software allows the digital reassembly of the bag voxel-by-voxel to form a complete 3D image, where the density and effective atomic number (known as Z-effective or simply Zeff) of each voxel is obtained, whilst also providing an accurate assessment of every object’s precise size and mass. Further, the 3D image provides much greater – and more intuitive – information than AT’s often cluttered, 2D projection views of the bag contents.

It does this by rotating the X-ray system around the object and producing a number of cross-sectional slices through the bag, which are processed by a computer to generate a three-dimensional representation of the object. A simple way to think about a CT scan of a bag is to imagine cutting a loaf of bread into slices and then cutting each slice into small cubes that can later be reconnected together to re-form a complete loaf. A CT scanner, with its rotating ‘X-ray camera’, is electronically doing the slicing and dicing. Each slice consists of a collection of small three-dimensional cubes, called ‘volume elements’ or voxels, so every item in the slice is quantified in terms of material characteristics, such as density and dual energy (Zeff) photon absorptive properties, which are obtained by comparing the low energy content to the high-energy photon content of each voxel. Thus, for each point in space in the bag, there exists a measurement of the material characteristics of each voxel that can be independently analysed. All the voxels that have the same density and that are touching – both within the slice and between different slices – are reconstructed into an object. Since we know, by design, the size of each voxel, the system can calculate the mass or weight of that object (mass = density x volume) and, if dual energy detectors are used, the material’s effective atomic number (Zeff) can be utilised as an orthogonal discriminator to further improve detection while also reducing the number of false alarms generated. Using this information, a detection algorithm can compare each object against the properties of explosives to determine not only whether an object has the same density and Zeff as an explosive on the list, but also more accurately determine if it is larger than the specified minimum mass. All of these factors improve detection while maintaining a very low false alarm rate.

While threat detection is the most important aspect of security, a low false alarm is critical so that each alarm has enough time to be resolved; the use of a dual-energy CT system can significantly reduce the number of false alarms produced. Further, the 3D images produced will reduce the amount of time an operator requires to correctly perform on-screen resolution and, if needed, a hand search. A lower false alarm rate also minimises the number of passengers subjected to such a search, which also infringes upon their privacy and convenience.

CT vs. AT

Historically, multi-view X-rays, being faster and lower cost, were originally used for checked baggage. But since 9/11, they have been – or are being – replaced with CT-based solutions.

The challenge at the checkpoint is more severe than hold baggage. Not only must explosives (not necessarily configured as IEDs, as they can later be assembled after screening) be detected, but other restricted items, like guns and sharps, are of concern as potential hijacker weapons. With 3D-capable CT, an operator can, using the software tools available, digitally inspect a bag’s contents from any angle to find non-explosive prohibited items while clearing a bag that contains only innocuous items and materials. Essentially, the CT is allowing the operator to ‘electronically unpack’ the bag. If a threat or restricted item is identified, the 3D image allows a searcher to rapidly and easily locate it within the bag.

Most manufacturers have acknowledged that AT systems are not capable of passing the ECAC C-3 test for liquid-born threats or C-4 advanced explosives threats test on their own. As a result, most are developing CT systems for future checkpoints. CT technology has been proven, with hold baggage screening, to be a logical approach to automatically and rapidly detecting explosives and presenting effective 3D images that operators can use to accurately examine 100% of the bag contents and locate any restricted items.

A laptop, presumed to be similar to the one used in the attempted bombing of Daallo Airlines Flight 159 in Somalia in 2016
Fig. 2: A laptop, presumed to be similar to the one used in the attempted bombing of Daallo Airlines Flight 159 in Somalia in 2016.

One example of a recent challenge facing airport security professionals is that of explosives, hidden within electronics, such as laptop computers and other PEDs. Figure 2 shows the damage done to Daallo Airlines Flight 159 in Somalia; the aircraft had not reached full pressurisation at the time of the explosion. The laptop shown was a second laptop rigged with explosives, that was captured before getting onto an aircraft (exploded afterwards for safety), and is presumed to be similar to the one used on the Daallo Airlines flight. Figure 3 represents a bag loaded with a similar laptop-borne explosive as seen on an X-ray scanner and a CT scanner. CT imaging captures these threats with unsurpassed accuracy.

  Shows the second laptop, presumed similar to the one that damaged Daallo Airlines Flight 159, as seen in an X-ray image view versus a CT image of the same bag.
Fig. 3: Shows the second laptop, presumed similar to the one that damaged Daallo Airlines Flight 159, as seen in an X-ray image view versus a CT image of the same bag.

When a threat is a well-hidden thin sheet explosive, it is extremely difficult, if not impossible, to detect with either traditional X-ray or multi-view X-ray scanners, while a CT scanner automatically detects these threats as it is analysing the bag from every angle.

In addition, since CT has accurate material data, it can be programmed to automatically detect weapons based first on their material density and then via shape analysis.

Shows how other more traditional restricted items can be easier to see on a CT image versus an X-ray or AT image in the presence of any clutter
Fig. 4: Shows how other more traditional restricted items can be easier to see on a CT image versus an X-ray or AT image in the presence of any clutter.

In conclusion, the use of CT systems at the checkpoint offers major advantages for automatic detection and image analysis. If cost and other deployment factors (such as power and weight) can be managed, CT offers the most proven way forward for cabin bag inspection, as it did for hold baggage. Once certified and deployed, checkpoint CT systems should also improve the passenger experience by eliminating the need for divesting electronics. Indeed, once such systems achieve ECAC C-3 certification, the restrictions on liquids, aerosols and gels (LAGs) may also be eliminated, speeding up the checkpoint screening process. Further, with the wealth of information collected on each bag, CT will allow the largest latitude for regulators and agencies to respond to intelligence regarding new threats. It will also protect the travelling public from new threats more effectively via software upgrades rather than requiring costly and time-consuming total system replacements.

Terrorist groups continue to plot attacks against aviation in order to cause mass casualties and significant economic damage, as well as to create media coverage. CT, when deployed at the checkpoint, will be a powerful tool for thwarting such attempts. It will also increase passenger throughput by eliminating the need to divest PEDs and LAGs, reducing false alarm rates and speeding up the secondary bag search process. In addition to better security, passengers will appreciate a simpler – and faster – security experience.

Joe Paresi
Joe Paresi

Joe Paresi is the founder, Chairman and Chief Executive Officer of Integrated Defense and Security Solutions, IDDS, which has developed the DETECT™ 1000. Prior to IDSS, he was the co-founder and Executive Vice President of L-1 Identity Solutions, Inc. Prior to L-1, Joe Paresi served as Corporate Vice President of Product Development at L-3 Communication Corporation and President and founder of L-3 Security & Detection Systems, where he led the development and deployment of the L-3 eXaminer 3DX 6000 TSA Certified Explosive Detection System and the ProVision millimetre wave body scanners. Mr. Paresi also served as Corporate Director of Technology for Lockheed Martin and Loral Corporations.