# Anti-loosening bolt looseness diagnosis method for flange node based on quick response code

## TL;DR

This paper introduces a method to detect bolt loosening in flange connections using QR codes and image correction techniques for improved accuracy and reliability.

## Contribution

A novel bolt image correction method using QR codes and a loosening diagnosis approach that reduces camera position and rotation effects.

## Key findings

- The proposed method effectively corrects perspective distortion in bolt images.
- It reduces the impact of camera position deviation and simultaneous rotation of screws and nuts.
- The method successfully detects loosening angles of anti-loosening bolts.

## Abstract

Flange connections are widely used in various fields such as civil engineering, aerospace, and mechanical manufacturing due to their convenient installation and reliability. However, the bolts in the flange connections may become loose, which could affect the safety of the connections. Nowadays, the vision-based detection techniques are widely used because of its ability to provide quantitative loosening angles, high efficiency, and low cost. However, the vision-based detection techniques are susceptible to the influence of perspective angle and it is hard to provide the historical information of bolt automatically. Therefore, the quick response code is introduce to record the historical information of bolt and a novel bolt image correction method is proposed based on quick response code. Furthermore, the anti-loosening bolt looseness diagnosis method is established. Firstly, the quick response code is designed with three finder patterns and pasted on the end face of screw. And then, the four intersections of quick response code edge lines are used to correct the image based on the homography-based perspective rectification method. Additionally, three finder patterns are used to rotate the image to the unified reference state to reduce the effects of camera position deviation and simultaneous rotation of screws and nuts. Finally, the anti-loosening bolt looseness diagnosis method is established by using the change in rotation angles of nut under initial status and loose status. A prototype flange node was used for experimental verification. The results show that the proposed method can effectively correct the perspective distortion of bolt image, reduce the effects of camera position deviation and simultaneous rotation of screws and nuts, and detect the loosening angle of anti-loosening bolts.