A first exploratory study on the relevance of everyday object knowledge and training for increasing efficiency in airport security X-ray screening

Abstract

Secure air transportation is vital for economy and society and it relies heavily on airport security screening. Passenger bags and other belongings are screened using X-ray machines to ensure that they do not contain prohibited items. Human operators (X-ray screeners) visually inspect X-ray images to decide whether they are harmless or whether they might contain a prohibited item and therefore require secondary search (typically using manual search and/or explosive trace detection technology). Several previous studies have shown that learning which items are prohibited and what they look like in X-ray images of passenger bags is important to achieve good detection performance. As passenger bags contain a large variety of harmless everyday objects, it could be assumed that knowing what such objects look like in X-ray images could help X-ray screeners to work more efficiently by reducing false alarms (i.e. sending a passenger bag to secondary search even though it does not contain a prohibited item). In the first experiment, the relationship between knowledge of harmless everyday objects and false alarm rate was investigated with 15 certified X-ray screeners of one large European airport. Statistical analyses revealed a good knowledge of harmless everyday objects on average with some variation between X-ray screeners and a negative correlation with false alarm rate. In the second experiment, the effectiveness of an e-learning course for acquiring knowledge of everyday objects in X-ray images was evaluated. Thirty novices conducted a test-retest experiment where half of the participants conducted an e-learning course about harmless everyday objects in X-ray images between the two tests. The results revealed that e-learning can be an effective and efficient method for increasing the knowledge of everyday objects in X-ray images. Based on the results of both studies, the relevance to learn everyday objects as part of initial and recurrent training of X-ray screeners is discussed.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 10 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.