Research On Digital Reference Image Of Digital Radiography For Weld Defects

X-ray digital radiography is the latest development in radiation detection and is widely used in non-destructive testing of pressure vessels,special equipment and aerospace components.Due to the different forms of welding defects and the difference in the level of the inspectors,there is often a controversy about the qualitative nature of the defects.Making a digital reference image of the weld defect rays is of great significance for the precise determination of the test results.In the process of X-ray DR imaging,affected by imaging equipment,imaging noise,etc.,it causes problems such as blurring and low contrast of DR images.It is necessary to study the enhanced display algorithm to enhance the display of defects in the image and improve the defect detection effect.Aluminum alloy lap welds were studied and the following studies were carried out:(1)Aiming at amorphous silicon flat panel detector imaging,from the perspective of digital image quality,three basic quality parameters of X-ray digital image are analyzed: contrast,image spatial resolution and signal-to-noise ratio.Taking aluminum alloy lap weld as the experimental object,X-ray DR detection was carried out,and the ray DR imaging inspection process of aluminum alloy lap weld was established by process test.According to the requirements of DR image quality in GB/T35388-2017 standard,the image quality of DR detection is evaluated.The results show that all aspects of the detection image meet the standard requirements.(2)DR imaging of aluminum alloy welds according to the inspection process,digital reference images of typical defects such as unfused,incomplete penetration,pores,inclusions,cracks and other defects are obtained,and the image features of various types of defects are described.According to HB 5375-1987 "Aluminum and Aluminum Alloy Welding Quality Inspection" standard and digital reference image,try to study the classification standard of defects,according to the single defect size and the total area of defects,the pore inclusion defects are classified into ?,?,?,Class ?.(3)For the problems of image blur and low contrast existing in the DR detection image,In order to improve the visual effect of defects,the image sharpening,the relief special effect display,and the multi-scale contrast enhancement method are used to enhance the display of defects in the image.The results show that the Laplacian sharpening method can improve the image sharpness and enhance the defects in the image,which is suitable for isolated point and line defects.Gradient direction edge enhancement and Prewitt edge enhancement are applied to the dots in the image.Defects and crack defects work well.The multi-scale contrast enhancement algorithm can effectively improve image contrast and improve defect detection effect.(4)Using Visual Studio 2010 as the development platform and object-oriented programming ideas,a set of weld ray DR image acquisition and evaluation software was developed in a modular design.The acquisition module is used for detector initialization,parameter setting,detector output correction,and DR image acquisition function implementation.The evaluation module is used to implement DR image loading,and realize image processing,region of interest measurement,Laplacian sharp The functions of the embossing and embossing effect display,the reference image module is used to implement the function of loading,classifying and displaying the defective digital reference image,and assisting the testing personnel to evaluate the detection result.