The integration of IT systems and data has evolved over decades. One of the earliest examples was in the banking industry, with the development of electronic data interchange (EDI), which allowed financial institutions to exchange data in specific formats between each other. Integration then expanded into the retail sector, and by the 1990s, it allowed businesses to connect their own systems to buyers and sellers worldwide, offering a clear competitive advantage.

Customs administrations soon joined the integration movement, with numerous projects being launched to improve data sharing and analysis at the national, regional and international levels. The digitalization of procedures, use of data analytics, and development of Single Window environments and of data exchange platforms in general, have all highlighted the benefits of data and systems integration. These integration initiatives have fostered international cooperation between governments, as well as nurtured public-private cooperation across borders.

Integration has also found application with regard to security systems, such as real-time monitoring devices, sensors and non-intrusive inspection equipment. Such technologies range from CCTV, to systems enabling licence plate recognition for vehicles, optical character recognition, radiation detection, under-vehicle surveillance, traffic management, automated temperature detection and radiography imaging.

Integration has particularly transformed the non-intrusive inspection (NII) equipment industry. In a first phase, NII systems were integrated into Customs management applications. The Original Equipment Manufacturers (OEMs) that could provide such integration could do so for their own systems and for the images generated by their equipment in a brand-specific format.

The concept of centralization of X-ray image analysis and interpretation then emerged, with remote screening centres replacing more traditional local processes where image analyses were adjudicated at the port of entry where the NII equipment was deployed. The first administrations to undergo such integration and centralization benefited highly from more automated processes and better utilization of human resources. However, these advantages were challenged when equipment from different suppliers was in operation, as each OEM delivered the radioscopic images in a proprietary format.

From then on, not only did OEMs grapple with the issue of integrating data generated by the equipment of different suppliers, but other companies started offering services as independent integrators. Once the integration of the data was accomplished, universal platforms or common viewers had to be designed to enable real-time remote image analysis of X-ray images generated by multiple systems and diverse brands. The last decade has seen the proliferation of such tools, developed both by independent system integrators and by several OEMs. These universal viewers provide a one-stop solution to gather, harmonize and analyse radiographic images generated by devices from different manufacturers. As new OEMs appear on the market with their own formats of the radiographic images, the circle continues, with system integrators needing to integrate new image formats.

These complex integration architectures, however, often lack consideration of the training aspect. Customs administrations should also consider their training needs and include a remote training platform capability into their integration architecture. Indeed, radiography image analysis is a challenging task, and enhancing operator capacities[1] through training is essential. The objective of the training is to enhance the detection performance of the screener, as well as to reduce the reaction time required for recognition to occur. One key question in image analysis training is whether screening officers should be trained using real or synthetic images, or both. While each approach has its merits, what is most important is to have access to a huge number of images.

One way to ensure access to more images is to integrate radiographic systems into a centralized platform and, additionally, into a computer-based training platform. With such integration, Customs administrations can ensure constant flows of data from their own operations, while at the same time have access to thousands of other images for training purposes.

Most importantly, they could benefit from having tailored and specific X-ray libraries, developed based on their needs and according to the operational challenges they face. However, establishing such libraries requires the insertion of clear radiographic images, the radiographic images of objects of interest. Only when such a merge is done using the raw format of the images can we obtain realistic scenarios. As each OEM uses proprietary image formats, system-agnostic training providers have to work with a myriad of raw X-ray image formats, just like system integrators.

The WCO’s Technical Experts Group on Non-Intrusive Inspection (TEG-NII) has further supported the integration of NII systems by developing a unified X-ray file format for high energy non-intrusive inspection devices, codenamed the Unified File Format (UFF).

Once deployed on all NII equipment, the UFF significantly facilitates the interoperability of NII equipment supplied by different manufacturers, as well as the exchange of images within and between Customs administrations.

In those countries which have already purchased NII scanners that use the UFF, or which have been able to have the UFF deployed on machines they already had in use, the integration of X-ray equipment into a remote command centre, and furthermore, the integration of radiographic images into a training platform, has been greatly simplified. The standard has been instrumental in developing the large databases or libraries of images that are, inter alia, necessary to train analysts, and in enhancing adaptive computer-based training (CBT).[2]

We therefore encourage Customs administrations which are considering centralizing scanning operations to make use of the UFF and to consider including a remote training platform in their integration architecture.

More information
www.casra.ch
info@casra.ch

[1] Michel, S. et al. (2014), Increasing X-ray image interpretation competency of cargo security screeners, International Journal of Industrial Ergonomics, Vol. 44(4), pp. 551-560.

[2] A method is called adaptive if it is able to select an item on the basis of trainee performance at an earlier stage. The term CBT is used for activities in which the computer presents a problem, registers answers and provides feedback. See The challenges of X-ray image analysis and the value of training – WCO.