A low cost visual monitoring system and method for rail bolt loosening

CN122156299APending Publication Date: 2026-06-05WUHAN JIAYING INTELLIGENT TECH CO LTD

Patent Information

Authority / Receiving Office: CN · China
Patent Type: Applications(China)
Current Assignee / Owner: WUHAN JIAYING INTELLIGENT TECH CO LTD
Filing Date: 2026-02-10
Publication Date: 2026-06-05

AI Technical Summary

Technical Problem

Existing railway bolt loosening monitoring technologies suffer from problems such as difficulty in monitoring coverage, high cost, poor environmental adaptability, and low detection stability, making it difficult to achieve widespread application.

Method used

An image acquisition module consisting of multiple PTZ cameras is used, combined with an edge device processing module, a local area focusing enhancement module, a global multi-scale scanning module, and a spatiotemporal sequence alignment scanning module. Bolt detection and loosening calculation are performed using the YOLO-Pose model and deep learning algorithms, achieving multi-view, low-cost, and high-precision monitoring.

Benefits of technology

It enables multi-view, large-area, full-coverage monitoring of railway bolts, reducing equipment and maintenance costs. It has sub-pixel-level detection accuracy, can operate stably in complex environments, has a processing delay of less than 1 second, supports rapid alarm notification, and improves the availability and scalability of the system.

✦ Generated by Eureka AI based on patent content.

Smart Images

Figure CN122156299A_ABST

Patent Text Reader

Abstract

The present application relates to a kind of low-cost visual monitoring systems of rail bolt loosening, including image acquisition module, edge device processing module, local area focus enhancement module, global multiscale scanning module, space-time sequence alignment scanning module, cloud platform module and communication module, the image acquisition module is made of multiple ball machine deployed along railway, for multi-angle acquisition bolt image, the edge device processing module is deployed in field edge computing device, for image pre-processing, bolt detection, key point extraction and loosening calculation, the present application realizes low-cost, high-precision, multi-view monitoring to railway bolt loosening, effectively solve the problems, such as the monitoring coverage difficult, high cost, poor environmental adaptability, low detection stability of prior art, provide reliable guarantee for railway operation safety.

Need to check novelty before this filing date? Find Prior Art