Automated Flaw Detection for NDE Images
Ye Tian, Amgen Inc., Thousand Oaks, CA 91320; Ranjan Maitra, Department of Statistics and
Statistical Laboratory, Iowa State University, Ames, IA 50011; William Q. Meeker, Department
of Statistics and the Center for NDE, Iowa State University, Ames, IA 50011; Stephen D.
Holland, Department of Aerospace Engineering and Center for NDE, Iowa State University,
Ames, IA 50010
Many modern NDE systems generate image data. In some applications an experienced
inspector performs the tedious task of visually inspecting every image to provide accurate
conclusions about the existence of flaws. This labor-intensive approach can cause misses due to
operator error. Automated methods can eliminate human-factors variability and improve
throughput. Simple methods based on peak amplitude in an image are sometimes employed and
a trained-operator-controlled refinement that uses a dynamic threshold based on signal-to-noise
ratio (SNR) has also been implemented. We develop an automated and optimized detection
procedure that mimics the operations of a trained operator. The primary goal of our
methodology is to reduce the number of images requiring expert visual evaluation by filtering
out images that have strong evidence for the existence or absence of a flaw. We use an
appropriate model for the observed values of the SNR-detection criterion to estimate POD. Our
methodology outperforms current methods in terms of its ability to detect flaws.
This work was performed with partial support from the Federal Aviation Administration
under contract number DTFACT-09-C-00006 through the Center for Nondestructive Evaluation
at Iowa State University.
Experimental Studies on NRUS Behavior of Impact Damaged Composite Laminates
Hyunjo Jeong, Wonkwang University, Mechanical and Automotive Engineering, Iksan, Jonbuk,
Korea; Daniel Barnard, Iowa State University, Center for NDE, Ames, IA 50011
Nonlinear acoustic methods or Nonlinear Elastic Wave Spectroscopy (NEWS) techniques
are widely used these days for damage assessment in a large group of materials including metals,
rocks and composites. Nonlinear resonant ultrasound spectroscopy (NRUS) is one of those
techniques and has proved to be valuable for damage detection because of its high sensitivity.
NRUS is a vibration-based technique exploiting the nonlinear resonance behavior of damaged
materials. In NRUS, the resonant frequency of an object is sought as a function the excitation
level. As the excitation level increases, the material nonlinearity is manifested by a shift in the
resonance frequency. In this work, NRUS experiments were conducted on intact and impactdamaged carbon fiber-epoxy composite laminates. Both fundamental and higher harmonic
resonance were measured and analyzed. It was observed that undamaged samples exhibited some
baseline nonlinearity. With increase in impact damage, nonlinearity increased beyond the
baseline nonlinearity. The hysteretic nonlinear parameter was extracted from the change in
resonant frequency with peak strain amplitude. The higher order hysteretic parameters were
found to be more sensitive to conventional first order parameters for given damage level.
This work was supported by National Research Foundation of Korea (2013R1A2A2A01016042).