Analysis of order-statistic CFAR threshold estimators for improved ultrasonic flaw detection

Abstract

In the pulse-echo method using broadband transducers, flaw detection can be improved by using optimal bandpass filtering to resolve flaw echoes surrounded by grain scatterers. Optimal bandpass filtering is achieved by examining spectral information of the flaw and grain echoes where frequency differences have been experimentally shown to be predictable in the Rayleigh scattering region. Using optimal frequency band information, flaw echoes can then be discriminated by applying adaptive thresholding techniques based on surrounding range cells. The authors present order-statistic (OS) processors, ranked and trimmed mean (TM), to robustly estimate the threshold while censoring outliers. The design of these OS processors is accomplished analytically based on constant false-alarm rate (CFAR) detection. It is shown that OS-CFAR and TM-CFAR processors can detect flaw echoes robustly with the CFAR of 10 (-4) where the range cell used for the threshold estimate contains outliers.

Real-time FPGA implementation of a reconfigurable ultrasonic detection system

“…Common flaw/target detection algorithms solely depend on the fact that clutter echoes in target materials exhibit randomness and are more sensitive to different frequencies in comparison with the flaw echoes [1][2][3]. Using this concept, algorithms such as Split-Spectrum Processing (SSP) can be used to distinguish the flaw echoes from the clutter echoes by studying the echoes in different narrow frequency bands [4][5].…”

Dynamically Reconfigurable Architecture Design for Ultrasonic Imaging

“…An important question is determining the frequency bands (scales) to be used for postprocessing. Since the clutter echo spectrum is shifted toward higher frequencies [8], the target echo is expected to be the dominant information in lower frequencies. Wavelet domain scales in Fig.…”

“…In the ultrasonic imaging of materials, an effective method of obtaining frequency diverse information is through split spectrum processing (SSP) of the broadband echoes [1], [8]. SSP can be implemented via subband signal decomposition.…”

“…The received signal is then digitized and passed through several bandpass filters using forward and inverse fast Fourier transforms (FFTs) to split the spectrum into different subbands (i.e., observation channels). The output signals from the subbands are then passed into a postdetection processor for target detection [1], [8]. This processor can employ different techniques such as signal averaging, Bayesian classifiers, and order statistic filters, including minimization.…”

Reconfigurable Architecture for Ultrasonic Signal Compression and Target Detection

“…Target detection algorithms are based on the premise that clutter echoes exhibit randomness and are more sensitive to frequency shifts than target echoes [1]. Therefore, frequency diverse signal decomposition such as discrete cosine transform (DCT), and discrete wavelet transform (DWT) can be used for differentiating the target information from the clutter echoes.…”