Abstract
This paper describes an experimental trial for detection of defects (or special features) on a pipe inner wall using a laser-beam scanning sensor. A strategy in which two simple threshold levels are set for the sensor outputs is proposed. If either threshold is exceeded, the sensor classifies the corresponding points as defective. There is, however, an epicenter caused by the difference between the center of the sensor and the center of the measured pipe. Therefore, the sensing system easily confuses actual defects and the signal caused by the epicenter. To avoid this problem, a digital notch filter is derived by using a series of three points measured while scanning around the inner circumference of the pipe. This filter computationally eliminate the epicenter. To determine the effectiveness of the algorithm with the thresholds and the filter, a model pipe that has some defects on its inner wall is constructed. The algorithm precisely detects the two holes as well as a disconnected coupler and three other couplers.
