Computer aided detection of crack regions in welded images using soft computing methods

Abstract

Welding is an essential component of many industries such as the newlineautomotive industry, the construction industry, the aviation industry, and newlinemore. Without this form of metal work, so many things, including many newlinebuildings, gates, and fences, small kitchen appliances, vehicles - and even newlinespace travel - would not exist. In general, though there are numerous types of newlinewelding available, four types of welds are frequently seen in many materials. newlineThey are Good weld case, Excess weld case, No weld case and Undercut weld newlinecase. In case of Good weld, the welding regions on the joined materials are newlinevery perfect and the weld is lasting for several years. newlineThe main contribution and originality of this thesis is to develop an newlineefficient and fully automated method for detecting defects in welded materials newlineusing soft computing approaches in order to improve the weld image newlineclassification accuracy. In this thesis, three different methods are proposed to newlinedetect and classify the weld images. newlineThe proposed method 1 develops newline