Study On Imaging Technology For Pipeline Magnetic Flux Leakage Inspection

Magnetic flux leakage (MFL) inspection is a commonly method on tube testing, which is a method that the adoption of MFL inspect to detect extent of damage or corrosion after magnetization wall of ferromagnetic materials. In the testing work that showed a clear shape of wall and location of the defect can be visual display pipeline status, and to improve the efficiency of inspection. In the field of practical application of MFL, this issue has important practical value and significance.This article is a study of the visual pipeline MFL inspection imaging technology. The methods that display wall and defects based on the magnetic signal information were studied and realized, including the contents of the MFL signal preprocessor, the quantitative of pipeline defect-characteristics of geometry, 3D reconstruction of the defective and the 2D pseudo-color Map of wall. Through the preprocessor so that the MFL signals into image, and presents the methods of quantitative recognition for the measure of defects base on MFL inspection. By the information of length and width of defects extraction from signal waveform, and mathematical models model from model recognition, this paper utilizes the potential function to estimate the depth of defects. Then the 3D reconstruction images of defects are created according to the result that quantitative of pipeline defects. Drawing 3D images using OpenGL, through 3D modeling, to deal with realistic, human-computer interaction, achieve a more intuitive 3D reconstruction. In addition, creation of digital maps base on MFL signal after preprocessed, clear visual display wall of the overall situation.The pipeline MFL imaging systems is developed on the platform which is based on Windows XP system and Visual C++6.0. The MFL signal be input for the system through text, and output Geometric features of defects after analyses of Algorithm, and Shows that the pseudo-color map of pipeline wall and three-dimensional images of defects. The system can also input training sample of defects for training potential function, results will be in the form of text stored in the hard disk. The experiment results prove that verification of the effectiveness of the system, all functions to meet the design requirements.