Location

Brunswick, ME

Start Date

1-1-1992 12:00 AM

Description

Detection and enhancement of low contrast flaws in radiographic images with high noise fields is an ongoing topic of research in nondestructive evaluation. In film radiography, the minimum detectable flaw thickness is controlled by the exposure characteristics and the flaw size in relation to the thickness of the part. The exposure characteristics determine the overall sensitivity and noise level, while the flaw thickness controls the contrast of the flaw image with respect to the background film density. Often it is difficult to generate optimal exposures when inspecting thick objects or complicated part geometries. This can result in noisy images due to the poor counting statistics of the photons as well as optical film densities that are suboptimal for visual interpretation. In addition, flaw contrast is often extremely low due to the flaw size or the poor orientation of crack-like flaws. The goal of the work presented in this paper is to demonstrate the utility of digital image processing using the matched filter for detecting and enhancing flaws in low-contrast, high-noise radiographic images. The basic theory of the matched filter will be presented along with its application to two-dimensional images, In addition several practical examples will be shown on simulated and real radiographs.

Book Title

Review of Progress in Quantitative Nondestructive Evaluation

Volume

11A

Chapter

Chapter 3: Interpretive Signal Processing and Image Reconstruction

Section

Signal Processing

Pages

879-886

DOI

10.1007/978-1-4615-3344-3_113

Language

en

File Format

application/pdf

Share

COinS
 
Jan 1st, 12:00 AM

Application of Two-Dimensional Matched Filters to X-Ray Radiographic Flaw Detection and Enhancement

Brunswick, ME

Detection and enhancement of low contrast flaws in radiographic images with high noise fields is an ongoing topic of research in nondestructive evaluation. In film radiography, the minimum detectable flaw thickness is controlled by the exposure characteristics and the flaw size in relation to the thickness of the part. The exposure characteristics determine the overall sensitivity and noise level, while the flaw thickness controls the contrast of the flaw image with respect to the background film density. Often it is difficult to generate optimal exposures when inspecting thick objects or complicated part geometries. This can result in noisy images due to the poor counting statistics of the photons as well as optical film densities that are suboptimal for visual interpretation. In addition, flaw contrast is often extremely low due to the flaw size or the poor orientation of crack-like flaws. The goal of the work presented in this paper is to demonstrate the utility of digital image processing using the matched filter for detecting and enhancing flaws in low-contrast, high-noise radiographic images. The basic theory of the matched filter will be presented along with its application to two-dimensional images, In addition several practical examples will be shown on simulated and real radiographs.