A random matrix approach to detect defects in a strongly scattering polycrystal: how the memory effect can help overcome multiple scattering

TL;DR

This paper introduces a random matrix method leveraging the memory effect to distinguish single from multiple scattering in ultrasonic imaging, enabling detection of deep flaws in polycrystals where conventional methods fail.

Contribution

The study demonstrates how the memory effect can be exploited with a random matrix approach to improve flaw detection in strongly scattering polycrystalline materials.

Findings

Flaws detected beyond conventional imaging limits.

Memory effect helps separate single and multiple scattering.

Enhanced ultrasonic imaging in complex media.

Abstract

We report on ultrasonic imaging in a random heterogeneous medium. The goal is to detect flaws embedded deeply into a polycrystalline material. A 64-element array of piezoelectric transmitters/receivers at a central frequency of 5 MHz is used to capture the Green's matrix in a backscattering configuration. Because of multiple scattering, conventional imaging completely fails to detect the deepest flaws. We utilize a random matrix approach, taking advantage of the deterministic coherence of the backscattered wave-field which is characteristic of single scattering and related to the memory effect. This allows us to separate single and multiple scattering contributions. As a consequence, we show that flaws are detected beyond the conventional limit, as if multiple scattering had been overcome.