Image processing-based magnetic particle inspection defect intelligent identification detection system

Abstract

The invention discloses an image processing-based magnetic particle inspection defect intelligent identification detection system. A communication module, a CCD (Charge Coupled Device) camera, a CCD control module, a crack identification module, an image processing module, an image alarm terminal and a memory are integrated in the system; a surface image of a workpiece is acquired by the CCD camera and is stored in a memory card; the image processing module is used for segmentation, defect identification processing and storage of the image in the memory card and transferring defect information of the surface of the workpiece to the crack identification module; and the defect information of the surface of the workpiece is fed back to the image alarm terminal by the crack identification module to finish the final display alarm. By improving a segmentation algorithm, the problems such as imaging fuzziness caused by motion, imaging region selection and false alarm caused by pseudo-crack signal existing in the system are solved excellently.

Description

technical field [0001] The invention relates to the field of non-destructive testing of magnetic particle flaw detection, in particular to an image processing-based intelligent identification and detection system of magnetic particle flaw detection defects. Background technique [0002] At present, the fully automatic magnetic particle flaw detection technology that uses a microcomputer intelligent system to replace human eyes to judge defects is in the research stage, and no mature system has been launched yet. [0003] The fluorescent magnetic particle automatic identification system developed by Beijing University of Technology has acquisition, smoothing, enhancement, and display modules, but no actual identification system has successfully run on the inspection production line. [0004] Some non-destructive testing manufacturers in China have launched trial products. The intelligent fluorescent magnetic particle flaw detector developed by the Second Department of the Sec...

AI Technical Summary

Problems solved by technology

[0009] However, the complexity of the surface image of the workpiece and the weak signal of the crack make this research have many difficult problems to solve:

The complexity of the image of the train wheel pair workpiece, the characteristics of weak signals and the difficulty of controlling the image quality make the detection algorithm very complicated, and it is difficult to process the data in real time during the data collection process to obtain the results

Images