Computer Imaging for the Electromagnetic Inspection of Oil-Country Tubular Goods

Abstract

Summary A new imaging system has been developed for the nondestructive evaluation of oil-country tubular goods (OCTG). It consists of smallarea-resolution magnetic sensors, data acquisition, digital signal processing, and a computer for image processing and pattern recognition. The main output of the system is a map of the pipe showing defects at the proper locations and orientations. Information on signal amplitude and waveform is also displayed. The system thus provides, in addition to the traditional information on signal waveform, information on contiguity and geometrical distribution of defect signals. Detection performance is enhanced by taking full advantage of digital signal processing. Depending on inspection needs and custom requirements, various signal-processing algorithms can be substituted at will by just loading new signal-processing programs. This flexibility, combined with small sensors, allows simulation and simultaneous use of many conventional detector configurations. The system described is the first step toward fully automated intelligent systems. It automatically extracts from the display image the length and orientation of suspected defects to produce a prove-up summary that includes longitudinal and circumferential position. This summary can expedite and refine the process of verifying the defects detected by the electronic inspection.