arXiv.itsnick.co

an independent field guide to 23 million papers from arXiv & OpenAlex — not affiliated with arXiv.org

Computer Science · Computer Vision and Pattern Recognition

Wavelet‐based blind deconvolution of near‐field ultrasound scans

Jason Taylor, Jose J. Mijares Chan, Gabriel Thomas

· IET Image Processing

See on the map

Abstract

A wavelet‐based technique for blind deconvolution and denoising of ultrasound scans is introduced. The target application is near‐field ultrasound imaging for non‐destructive testing. Existing blind deconvolution techniques for ultrasound such as cepstrum‐based methods and the work of Adam and Michailovich – based on discrete wavelet transform (DWT) shrinkage of the log‐spectrum – estimate the pulse by exploiting the pulse log‐spectrum smoothness relative to the material reflectivity function. In the proposed technique, the log‐spectrum is localised with respect to time as the continuous wavelet transform (CWT) log‐scalogram to deal with the non‐stationarity of the near‐field ultrasound signals in both the pulse estimation and deconvolution. The pulse is estimated in the wavelet domain via DWT shrinkage of the log‐scalogram and is deconvolved by wavelet‐domain Wiener filtering. Extensions of the proposed technique include: using separate CWT domains for estimation and deconvolution, as inspired by the WienerChop denoising method; and training the algorithm parameters on a subset of scans.

More like this