Study On Weld Defect Model And Classification Algorithm Of X-ray Submerged Arc Welding

With the development of image processing and pattern recognition technology,computer intelligent evaluation sheet,with its advantages,such as high efficiency,objectivity and economy has been widely applied to oil and gas pipelines in the field of weld defect detection.However,the traditional recognition methods require plentiful number of training samples,and it leads lots problem of running time.According to this,a method requires small amount of training samples is proposed in this research to classify different welding seam defects of oil welded pipe,it not only distinguish the welding defects from image noise,but can tell the specific category.It will helpful in the further improvement on the welding procedure.This article adopts the weld defect detection platform to identify the welding defects,such as: circle defects and linear defects,which may exist in the welding seam,a process includes: image processing,feature description and classification recognize.First of all,through image filtering the results of the analysis type of weld image noise and image after enhancement experiment,select appropriate image processing algorithms to remove noise,then the Sin enhancement,OSTU segmentation and sobel method is used to find the welding seam boundary in the entire image,and then the Hough transform is adopted to calculate the related parameters of boundary line,in order to achieve the ROI region segmentation.And then,by comparing the result of Ostu algorithm and density clustering for weld defects and segmentation of image noise,and then choose the latter way to split the ROI region.Secondly,traditional methods were used to extract the geometric parameters and extract characterization of pixel gray circular of circle defects and linear defects to describe,to establish the appropriate characterization of vectors,and use the Principal Component method preprocess this vector,and then we decide using pixel gray features to describe.Finally,studied on the result of the recognition of characteristic parameters of the classical SVM and fuzzy C-means clustering algorithm,of which the processed vector is taken as the input parameter,the final model were built based on the experimental results and we developed a weld defects automatic detection system based on RAD Studio,which realized the accurate identification of weld defects.Multi-core paralel programming techniques were used aim at reducing the running time of defects recognition process.Finally,this method can satisfy therequirements of real-time.