Automatic recognition of truck chassis welding defects using texture features and artificial neural networks 2019-01-1119
Welding is an excellent attachment or repair method. The advanced industries such as oil, automotive industries, and other important industries need to rely on reliable welding operations; collapse because of this welding may lead to an excessive cost in money. In this paper, an automatic system to detect, recognize and classify welding defects in radiographic images was described using texture feature and neural network techniques. Image processing techniques, including converting color images to grayscale, filtering image, and resizing were implemented to help in the image array of weld images and the detection of weld defects. Therefore, a proposed program was build in-house to automatically classify and recognize eleven types of welding defects met in practice. The proposed system has been tested on eleven welding defects which are: center line crack, cap undercut, elongated slag lines, lack of interpass fusion, lack of root penetration, lack of side wall fusion, misalignment, root crack, root pass aligned, root undercut, and transverse crack. It was found that between two cases are failed in a total of 2 from 154 images, and 3 from 308 images, in training, and the overall classification percent, respectively. The lowest classification percent was found in case of lack of side wall fusion defect (92.9%). The overall average discrimination rate results from the combined approaches are about 99%.
Saeed A. Al-Ghamdi, A S Emam, Ossama B. Abouelatta
Albaha University, Helwan University, Mansoura University