Baggage Inspection System

The system addresses orientation variability in baggage inspection by rotating images and using machine learning to enhance accuracy and reduce inspector effort, improving throughput and reducing errors in baggage inspection systems.

JP7727451B2Active Publication Date: 2025-08-21NIPPON SIGNAL CO LTD
View PDF 13 Cites 0 Cited by

Patent Information

Application Number
JP2021145713
Authority / Receiving Office
JP · JP
Patent Type
Patents
Current Assignee / Owner
Filing Date
2021-09-07
Publication Date
2025-08-21
Estimated Expiration
2041-09-07

AI Technical Summary

Technical Problem

Existing baggage inspection systems using machine learning models for object recognition in X-ray images face inaccuracies due to varying orientations of baggage during inspection, leading to insufficient determination results.

Method used

A baggage inspection system that rotates inspection images at multiple angles and uses machine learning models to determine the presence of objects, restricting baggage passage based on these determinations, and displays only images confirmed to contain objects to inspectors, reducing visual inspection effort.

Benefits of technology

Enhances accuracy of object detection in baggage inspection and reduces inspector workload by minimizing the need for visual verification of all images, thereby increasing throughput and reducing errors.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 0007727451000001
    Figure 0007727451000001
  • Figure 0007727451000002
    Figure 0007727451000002
  • Figure 0007727451000003
    Figure 0007727451000003
Patent Text Reader

Abstract

To provide a hand luggage inspection system that uses a machine learning model bringing a precise determination result higher than a conventional art.SOLUTION: A hand luggage inspection device 11 is configured to: generate an X-ray transmission image of hand luggage as an inspection image; generate a plurality of rotating images to be obtained by rotating the inspection image around an axis in a normal direction of an image surface; and determine whether, as to the inspection image and each of the plurality of rotating images, the image includes an image of a hazardous material by a machine learning model. The hand luggage inspection device 11 is configured to transmit a determination result to a terminal device 16 an inspector uses. The terminal device 16 is configured to display the determination result received from the hand luggage inspection device 11 and transmit a command of opening / closing a gate to gate devices 13 to 15 in accordance with the determination result. The gate devices 13 to 15 are configured to open / close the gate according to the command received from the terminal device 16. As a result, a passenger, who carries hand luggage having the inclusion of the hazardous material determined, is guided to a lane R2 and gets inspected by the inspector.SELECTED DRAWING: Figure 1
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Airport passenger security check intelligent assistance system and method based on X-ray machine image

    CN111665572A

  • Automatic baggage sorting system

    JP2001233440A

  • Image processing apparatus, image processing method, program, and computer readable storage medium

    JP2010072826A

  • Security system, and hand luggage inspection device

    JP2018115969A

  • Inspection system

    JP2019191083A