基于X射线图像处理的石油焊管缺陷检测技术-检测技术与自动化装置专业论文.docxVIP

英文摘要Subject： 英文摘要 Subject： The Detection of oil pipe based on the X-ray image processing Speciality： Detecting Technology and Automation Equipment Name： Zha)(iaochun(si印atu旧．出逝丛坠 Instructor： Tang Nan(signature As the main factors，the weld defects affects the producing quflity of oil pipe and has become one of the main objectives of the image processing．This paper studies a variety of image processing technology in the oil pipe weld detection,and uses visual Delphi software on a variety of weld image processing． Due to the impact of the noise，the weld image，which is captured by the X—ray imaging system，has the characteristic of lower contract and vaguer edge．So it is difficult to analyes its image and identify its defects． Based on analyzing the series features of the weld image，some processing on the image has been used in operation,such as noise filtering，image enhancement，edge detection and feature extraction operation．First，the image noise has been filtered by the improved median filter algorithm，it improved the overall gray value of the image；Second，in the image enhancement，through the comparing of nonlinear sine function and neural network algorithm，it verified that the sine function enhancement algorithment Carl enhance the contrast of the target and background；Thired，the sobel edge detection operator has been used to detect image edge weld．At the same time，in order to distinguish by the computer,the discontinuous edge is fittinged by the least squares．Finally,the recursive binary image component labeling algorithm has been used to calculate the number of connected domains in the line of the internal defects of processed binary image edge．Through the analysis of defect characterication，we select a group parameters which can accurately reflects the characteristics of the defect nature，at one time，we give the calculation of the parameters．By the above method，we Call determine the type of the defect correctly． By a series of experiments，it is proved tha