Reconstruction of specular reflectors by iterative image source localization.

Area of Science:

• Acoustics
• Ultrasonic Imaging
• Non-Destructive Testing

Background:

• Accurate characterization of material interfaces is crucial in various NDT applications.
• Traditional ultrasonic methods face challenges in precisely reconstructing complex reflector geometries.

Purpose of the Study:

• To develop and validate a novel ultrasonic method for reconstructing specular reflector geometry.
• To leverage the image source principle for enhanced accuracy in ultrasonic imaging.
• To assess the method's performance under varying noise conditions.

Main Methods:

• Utilizing an ultrasonic array to focus beams and emulate a point source.
• Applying sound source localization algorithms to analyze reflected waves.
• Employing the mirror equation for spherical reflectors to extract geometry.
• Developing an iterative scheme for reconstructing arbitrary reflector shapes.

Main Results:

• Accurate reconstruction of specular reflector position (error < 0.2 mm), orientation (error < 3°), and radius of curvature (error < 0.2).
• Robust performance in the presence of thermal and speckle noise (SNR(th) > -3 dB, SNR(sp) > 7 dB).
• Successful validation of the iterative scheme for arbitrary reflectors both numerically and experimentally.

Conclusions:

• The proposed image source-based ultrasonic method offers high accuracy for specular reflector geometry reconstruction.
• The technique demonstrates resilience to common noise sources in ultrasonic testing.
• The iterative extension enables versatile application to complex, non-spherical reflectors.