The baggage customs control systems at airports detect and process cases of baggage containing contraband, i.e., goods that are illegal to import or whose import failed to be declared for customs clearance, such as drugs, tobacco, endangered species, raw food material, or currency. Depending on the risk assessment of a particular arrival, the baggage customs control staff may decide to run all baggage from the arriving flight through non-intrusive screening equipment (X-ray/CT), which helps them to identify suspicious luggage, which is then considered suitable for manual inspection.
The main challenge for the customs operation is to reidentify that luggage after the passenger had picked it up from the carousel and proceeded toward the exit. Furthermore, the current state-of-the-art in luggage reidentification has several drawbacks:
- Tags must be manually placed in or on the suspect luggage and then removed before the traveller leaves the customs area following the manual inspection of their luggage to ensure they do not face another inspection the next time they travel with the same bag.
- The smugglers may realize their luggage has been tagged and remove the tags before entering the customs area, thereby hindering the reidentification and capturing process.
- While some airports apply radio-frequency identification (RFID) tagging for the customs reidentification of luggage, the RFID tags might damage the bag during their removal.
- The current process has a relatively high operational and maintenance cost due to the manual labor involved, as well as an environmental impact caused by the production and immediate disposal of tags.
With the increase in the volume of passengers arriving by air at inland border airports, the need for increased efficiency of baggage customs control continues to grow.
BAG-INTEL’s envisioned solution to support the customs teams in efficiently tackling these challenges will involve high-resolution cameras and robust AI, which will provide an end-to-end continuous reidentification of luggage. The solution will help ensure that all baggage is scanned and that the suspicious luggage is kept track of so that the customs staff can easily find it to perform a manual inspection.
The BAG-INTEL solution will comprise three main components:
- an AI-camera-based end-to-end continuous reidentification of luggage,
- new functionalities for enhanced scanning combining AI-based detection with X-ray absorption technologies for recognition of contraband concealed in the luggage, and
- a digital twin for BAG-INTEL system visualization and performance optimization designed for an operational context of an airport.
The main advantages of the BAG-INTEL solution include:
- a completely non-intrusive end-to-end reidentification (no physical tags placed on the luggage),
- no need for additional human resources,
- no delays in baggage arrival at the carousel (imaging done while luggage is in motion),
- increased effectiveness and efficiency (the staff will be able to focus only on pieces marked as suspicious),
- interoperability to be easily interfaced with legacy systems.
To further improve the effectiveness and efficiency of customs control, the project will—in addition to the reidentification solution—provide new, AI-supported functionalities improved for the risk assessment of the luggage. BAG-INTEL will include new, dedicated X-ray absorption analysis and object recognition, as well as an analysis of relevant external data. Furthermore, the project will provide a digital twin for visualizing, testing, and optimizing the developed solution for the operational context of airports. This will support the end-users in identifying the setup with the best performance under specific conditions. Finally, BAG-INTEL will also develop a sustainable business model and a deployment plan for the Pan-European uptake of project results.