Compressive Sensing Strategy on Sparse Array to Accelerate Ultrasonic TFM Imaging.

Area of Science:

• Materials Science and Engineering
• Nondestructive Testing
• Signal Processing

Background:

• Phased array ultrasonic testing (PAUT) using full matrix capture (FMC) is increasingly vital for defect characterization.
• The combination of FMC and total focusing method (TFM) yields high-resolution images but is data-intensive.
• Data volume in FMC-TFM is proportional to the square of the probe's elements, posing hardware and processing challenges.

Purpose of the Study:

• To investigate the application of compressive sensing (CS) for reducing data requirements in FMC-TFM.
• To evaluate the effectiveness of CS in accelerating TFM imaging and assess its impact on image quality.
• To explore the use of sparse arrays in conjunction with CS for further optimization.

Main Methods:

• Experimental data acquisition using FMC with phased array ultrasonic transducers.
• Application of compressive sensing (CS) algorithms for data compression and signal recovery.
• Comparison of TFM images generated from FMC, CS-compressed FMC (CS-FMC), and sparse CS data using contrast-to-noise ratio (CNR).

Main Results:

• CS methodology reduced data volume by up to 80% while achieving adequate FMC data recovery.
• CS-FMC produced TFM images comparable in quality to those obtained from full FMC data.
• Sparse arrays combined with CS improved reconstruction times by up to 11-fold and accelerated image formation by 6.6 times, with minimal image quality degradation.

Conclusions:

• Compressive sensing is a viable technique for overcoming the data-intensive nature of FMC-TFM.
• CS enables significant data reduction and accelerated imaging without substantial loss of defect characterization capability.
• The integration of sparse arrays with CS offers a promising pathway for more efficient and faster ultrasonic imaging.