Enabling near-array ultrasonic imaging with an apex-shifted Radon transform

Abstract

This paper presents an ultrasonic signal processing technique based on the apex-shifted Radon transform (ASRT) to mitigate near-array artifacts as a more selective filtering alternative to conventional time-gating strategies for the total focusing method (TFM). The ASRT-based image processing algorithm described here is used to selectively mute the direct P-waves and surface waves (collectively referred to as direct arrivals) by their geometry in the time–space domain of full matrix capture data before performing TFM, which are the sources of near-array artifacts. First, a mathematical formulation and intuitive explanation of the ASRT is provided. Then, the algorithm’s ability to improve near-array image clarity is demonstrated on a synthetic model of a steel region with two small inclusions near the array. Here, a significant reduction of the “dead-zone” artifacts is observed. Then, the algorithm’s performance is shown to be comparable when applied to experimental data obtained from an Inconel block with holes. From this investigation, the ASRT is concluded to be an effective tool for enabling ultrasonic imaging within a wavelength of the array.