skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Ultra-high speed residue processor for SAFT inspection system image enhancement. Final report, October 1983-March 1984

Technical Report ·
OSTI ID:5786457

The Phase-I feasibility study of residue number system (RNS) image processing for SAFT inspection has successfully determined that an advanced inspection system may be built using a correlation-reconstruction SAFT algorithm, implemented with RNS techniques and off-the-shelf electronic components. Images are reconstructed in a number theoretic transform domain with simple pointwise multiplication of the A-scan data volume by a custom point spread function (PSF), all in a highly parallel computational architecture. These methods also allow image enhancement to be easily performed for improved flaw visualization, and with negligible speed reduction. It has been determined that high resolution three dimensional flaw images may be generated and that a commercially viable product could result through development of a prototype real-time RNS processor. The hardware is expected to be made up of 100 nsec bit slice microprocessor components and large RAM storage units. Such a processor would provide the NDE community with a valuable new tool, that could generate significant improvements in real-time flaw visualization of critical components encountered in nuclear power plant inspection. Special inspection conditions, such as multiple reflections and crack tip detection in the presence of corner reflecting effects, may be overcome by modifying the functional form of the PSF. It is also possible that semi-automatic flaw classification could be included in the process without significantly adding to the computational burden. Based on the performance estimates of the Phase-I effort, this new image processing system has the potential to acquire and focus the equivalent of 145 A-scans per second, which translates into more than 1000 in./sup 3//min inspection rate for typical pressure vessel specimens.

Research Organization:
Sigma Research, Inc., Seattle, WA (USA). Seattle Lab.
OSTI ID:
5786457
Report Number(s):
NUREG/CR-4170; ON: TI85901284
Country of Publication:
United States
Language:
English