Font Size: a A A

Based On Image Processing Detection Of Pipe Weld Quality

Posted on:2011-07-08Degree:MasterType:Thesis
Country:ChinaCandidate:R BaiFull Text:PDF
GTID:2178360332457562Subject:Pattern Recognition and Intelligent Systems
Abstract/Summary:PDF Full Text Request
With industrial production toward automation and intelligent direction of development, computer vision and image processing technology in the quality inspection of industrial parts in widely applied. Straight welded pipe of automatic identification and quality inspection is the image processing in industrial production in the two application focus. Straight welded pipe's characteristics for the thesis put forward different recognition algorithm, and on this basis, developed a quality test of recognition system.Based on Image Processing Longitudinal weld quality detection system is the use of CCD camera image of weld image acquisition card via the analog images into digital images stored in the computer, then after the image preprocessing and image recognition features are targets, namely: the horizontal width of weld. Thesis includes four parts: image acquisition, image preprocessing, image recognition weld, experimental analysis of pipe quality testing. Image acquisition design a set of image acquisition devices. Image preprocessing focuses on the image geometry correction and addition to noise filtering algorithms and binary image conversion. Image identify weld is mainly of the image recognition algorithm, then according to the design of algorithms developed using VisualC++6.0 straight welded pipe weld quality detection system software. It is 90% accuracy with high accuracy. Experiments on the design of the system algorithm verification, successfully testing the pipe quality. It is 90% accuracy with high accuracy. Finally, experimental measurement accuracy of weld made detailed comparison and analysis, on how to overcome the impact of environmental factors, further improve the detection accuracy of proposed improvements.This system overcomes many disadvantages of artificial test weld to avoid human factors. Greatly improve the detection efficiency and high precision, fast. In fairness, accuracy, low cost and so there is a lot of advantages, which have very good development and application.
Keywords/Search Tags:Weld, Quality Inspection, Edge Detection, Denoising, Image Recognition
PDF Full Text Request
Related items