Weld Image Recognition Research And Application

SAW is a kind of widely adopted technology both in china and abroad. How to detect the fault of welding seam is a difficult problem that has not been solved yet. It has been affecting product quality and efficiency in the course of producing welding pipes. Almost all the oil and gas in our country is transported with spiral welding pipes. During transportation, pipe-leaking accidents caused by the poor quality of the steel pipes often happen, which bring about great loss to the property of the country and sometimes even cause personal injury or death. During the course of general detection, the operator must stare at the monitor all the time, which can easily cause eye fatigue. Those who have been doing this job for a long time can have their canthus membrane damaged that may cause short sight. All that can hurt workers greatly both physically and psychologically. Accordingly, the research of this subject will bring great benefit both socially and economically. Reasons of how the welding seam come into being are studied in this paper. According to these reasons, the shape of every fault is deduced. At the same time this paper also introduces a few methods which are used in the welding seam detection at present. The basic theory of image processing is introduced in this paper. A special image processing method for the welding seam image is mentioned based on the image processing and the x- rays image of the welding seam fault. This paper also gives a detail introduction on the automatic welding seam fault detecting system. That is X-rays system for detecting faults of spiral pipes, which has been designed by the author. The system has the following context: choosing image sampling card, 8253 counter design, speed sensor design, blowgun controlling card design, system module division and computer software flow chart. At the end of this paper, some running status of this system is demonstrated and some point which need to be improved further are pointed out.