Analytic evaluation of sampled aperture ultrasonic imaging techniques for NDE

This paper presents a theoretical comparison of three generic sampled aperture ultrasonic imaging systems for nondestructive evaluation. The imaging systems are categorized according to their source-receiver combination for data acquisition: common-source, back-scatter, and full-array imaging. First, forward modeling is performed for a point source and a point receiver. This is then used to model the received data set for each of the imaging categories. Subsequently, the inversion algorithm for each category is derived, and their performance is evaluated in terms of resolution, noise, and computation. We show that in terms of resolution, back-scatter imaging is the best, followed by full-array and common-source imaging. However, in terms of material noise, full-array imaging is the best, with back-scatter and common-source imaging having the same material noise response. Full-array imaging is the only system with inherent redundancy to reduce electronic noise, but at the expense of significantly more computation. The physical transducer is in the full-array category, allowing mechanical scanning to be traded for dynamic focusing and computational power.<<ETX>>