Using speed measures to predict performance in x-ray luggage screening tasks

A review of X-ray explosives detection techniques for checked baggage

“…However it was noted that both groups performed badly at detecting IEDs, even though the trained observer's eye dwelled on the threat, but invariably ignored it, underlining the need for appropriate training on this, the most difficult of threats for detection. This was also supported by other prior work (Wales et al, 2009), showing similar reductions when the detection task becomes more challenging (i.e. baggage image complexity increases), with poorest performance demonstrated for IEDs which do not exhibit such regular image-based features as other threat objects such as guns and knives.…”

“…TIP was planned to be a standard feature integrated into all Xray baggage inspection systems used in US airports by 2003. One of the standard methods for assessing baggage screener competency in the US was to subject screeners to the Competency Assessment Test (CAT), wherein a linear relationship has been found between hit rate and decision time (Wales et al 2009 prohibited and what items are accepted. It consists of a set of images of knives, sharp objects guns and explosives, and is designed to assess the knowledge component of a screener's skill set.…”

“…XRT delivers a graded training environment that increase with difficulty as the observer's performance during training increases (Mendes & Schwaninger, 2010). In the US, one of the standard methods for assessing baggage screener competency is to subject screeners to the Competency Assessment Test (CAT) (Wales et al, 2009). …”

A fuzzy model for evaluating airport security screeners' work

“…McCarley's (2009) paper points to the important role of any kind of aids, even minor ones, that indicate that special attention should be paid to a particular item of luggage in increasing the effectiveness of security screeners' work. In their article, Wales et al (2009) used the Threat Image Projection (TIP) system to assess a security screener's competence. A linear relationship between the response time and the number of images of prohibited items that had been detected was established.…”

Applications in X-ray Testing

“…Nevertheless, the 9/11 attacks increased the security measures taken at airports, which in turn stimulated the interest of the scientific community in the research of areas related to security using advanced computational techniques. Over the last decade, the main contributions were: analysis of human inspection [72], pseudo-coloring of X-ray images [73,74], enhancement and segmentation of X-ray images [75] and detection of threatening items in X-ray images, based on texture features (detecting a 9 mm Colt Beretta automatic (machine) pistol) [76], neural networks and fuzzy rules (yielding about 80 % of performance) [77], and SVM classifier (detecting guns in real time) [78].…”