Real-Time Detection of Irregular Baggage
Proactive Management of Even the Smallest Exceptions

Incheon International Airport Corporation has developed an artificial intelligence (AI)-based system for baggage anomaly detection and is set to patent the technology.


AI-Based Baggage Anomaly Detection System. Incheon International Airport Corporation

AI-Based Baggage Anomaly Detection System. Incheon International Airport Corporation

View original image

On September 7, the corporation announced that it had completed a joint development with POSCO DX of the ‘AI-Based Baggage Anomaly Detection System,’ which can dramatically reduce misclassification and damage incidents in baggage handling processes, and has filed for a patent on the technology.


Incheon Airport’s Baggage Handling System (BHS) already boasts world-class accuracy, with only two cases of unaccompanied baggage per one million items processed.


To deliver a flawless service that manages even the smallest exceptions, the corporation has developed an AI-powered detection system that can proactively identify rare occurrences such as double loading of baggage.


The newly developed technology is a smart airport solution that uses AI to analyze video footage captured by vision cameras. By applying a deep learning model trained on various baggage shapes, sizes, colors, and stacking patterns, the system provides real-time monitoring of baggage as it swiftly moves along the conveyor belt, precisely detecting anomalies that may cause system malfunctions or damages.


The main targets for detection are: ● ‘Double stacked baggage’—multiple bags overlapping and hindering barcode recognition; ● ‘Extended carrier handle’—a bag handle left open, posing a risk of snagging on surrounding facilities; and ● ‘Separation of baggage and tub’—cases where baggage is detached from its dedicated carrying tub.


The 'AI-Based Baggage Anomaly Detection System' instantly detects baggage exhibiting abnormal conditions and automatically diverts it to a manual processing area. This allows on-site personnel to re-inspect and handle flagged items safely, thereby preventing misclassification and damage.


The corporation and POSCO DX recently installed the system in Terminal 1 of Incheon Airport for pilot operation. The results proved the system’s outstanding detection capabilities, achieving a 100% success rate in identifying baggage anomalies that could cause system errors or damage.


Additionally, the two companies are developing a dedicated detection system specifically for oversized baggage. Once implemented, this will dramatically reduce the chances of unaccompanied items in oversized baggage handling and further enhance accuracy in this category.


Jin Hodon, Department Head at POSCO DX and a participant in the joint development, commented, “This system is a representative case where our Intelligent Factory technology platform—based on collaboration among people, AI, and robots—has been successfully demonstrated in an airport setting.”



Cho Yongsoo, Head of Operations at Incheon International Airport Corporation, stated, “With this patent application, we have secured unmatched technological capabilities to improve both the accuracy and operational efficiency of airport baggage handling. Going forward, we will continue to drive digital innovation throughout airport infrastructure by expanding the use of advanced intelligent technologies such as AI and the Internet of Things (IoT), enabling passengers to enjoy safer and more convenient airport services.”


This content was produced with the assistance of AI translation services.

© The Asia Business Daily. All rights reserved. Unauthorized AI training and use prohibited.

Today’s Briefing