Judgment system and judgment method

The determination system on a railway vehicle uses imaging and detection units to accurately assess abnormalities in objects like overhead wires and ridges, enhancing maintenance efficiency by reducing perspective errors and improving detection accuracy.

JP2026106935AActive Publication Date: 2026-06-30SUMITOMO CORPORATION
View PDF 1 Cites 0 Cited by

Patent Information

Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
SUMITOMO CORPORATION
Filing Date
2024-12-18
Publication Date
2026-06-30

AI Technical Summary

Technical Problem

Existing abnormality monitoring devices for shut-off valves in railway facilities are inefficient in performing maintenance work, and there is a need for a more accurate and efficient method to detect abnormalities in objects such as overhead wires and trees or excess ridges to facilitate timely maintenance.

Method used

A determination system mounted on a railway vehicle that includes an imaging unit to capture images of the surroundings, a detection unit to identify specific objects using pre-trained models, and a determination unit to assess abnormalities based on acquired object information, utilizing upward- and downward-facing cameras to reduce perspective errors and accurately determine proximity or height of objects.

Benefits of technology

Enables efficient and accurate detection of abnormalities in railway facility objects, allowing for timely maintenance by reducing perspective errors and improving maintenance efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 2026106935000001_ABST
    Figure 2026106935000001_ABST
Patent Text Reader

Abstract

To efficiently carry out maintenance work on objects within railway facilities. [Solution] The determination system 1 is mounted on a railway vehicle running on rails R1 and includes an imaging unit 21 that takes images of the area around the moving railway vehicle and outputs images of the target object; a detection unit 42 that inputs the images to a first trained model and detects at least one target object from the images; an information acquisition unit 43 that acquires information about the target object detected by the detection unit 42; and a determination unit 44 that determines whether or not there is an abnormality in the target object based on the object information.
Need to check novelty before this filing date? Find Prior Art