X-ray flaw detection intelligent detection method

Abstract

The invention discloses an intelligent detection method for X-ray flaw detection. The method comprises the following steps: carrying out real-time dynamic detection on a steel pipe by utilizing X-ray detection equipment; the computer receives the dynamic image in real time and sets the window level and the window width of the dynamic image; the detected dynamic image is stored in real time; establishing an intelligent defect identification model based on an artificial intelligence deep learning method; real-time dynamic image reasoning is conducted through the dynamic recognition model, an alarm is given after defects are detected, and a stop signal is sent to enable the steel pipe to stop moving; performing static image reasoning on the static steel pipe through the static recognition model, recognizing and uploading a defect image, and continuing dynamic detection of the steel pipe; and stopping the detection work and withdrawing the steel pipe after all-steel-pipe detection is finished. The method has the beneficial effects that a whole set of X-ray flaw detection solution is provided according to the radiographic flaw detection image of the welding seam, and the actual problem of a user site is well solved.

Description

technical field [0001] The invention relates to X-ray non-destructive testing and image intelligent recognition, in particular to an intelligent detection method for X-ray flaw detection. Background technique [0002] Due to the influence of factors such as different environmental conditions and welding processes, various defects (such as: cracks, pores, incomplete fusion, incomplete penetration, slag inclusions, etc.) will inevitably appear during the welding process, and due to the nature of the defects Some defects in the principle of radiography and radiography are relatively difficult to find, which makes the entire flaw detection process more complicated and makes film evaluation more difficult. These defects directly endanger the quality of welded products, and have a great impact on the safety and performance of welded structures. [0003] At present, the commonly used non-destructive testing methods mainly include ultrasonic, magnetic particle, ray, etc. Among them...

AI Technical Summary

Problems solved by technology

[0005] The existing manual evaluation and flaw detection technology has the following defects: 1. Due to the diversity of weld defect discrimination, full-time technicians observe the weld area in the X-ray image and perform comparative analysis to identify the defects in the image. Whether there are defects in the weld area, the manual evaluation method has a relatively subjective test result, and the evaluation is relatively flexible, which has great shortcomings

Different assessors may have different evaluation results on the same weld image, so this manual evaluation method will inevitably bring differences, which will affect the results of weld evaluation;

[0006] 2. Radiographic flaw detection is a highly professional non-destructive testing technical position, and the operator has high detection intensity and high pressure

[0007] 3. The dynamic flaw detection data of the pipe body is not archived, let alone re-inspection and traceability

Images