As an important component of railways, rails are prone to damage due to the stress exerted by train wheels and environmental factors. Shallow surface defects in rails pose a serious threat to train operational safety and present a major challenge in Ultrasonic Testing (UT). The boundary echo overlaps with the defect echo, hindering the extraction of defect signals from detection data. Therefore, this paper proposes a Phased Array Ultrasonic (PAU) imaging method based on Deconvolutional Delay-Multiply-and-Sum (DE-DMAS). In the preprocessing stage, a Wiener deconvolution-based ultrasonic echo processing method separates and extracts shallow surface defect echoes. In the beamforming stage, the Delay-Multiply-and-Sum (DMAS) algorithm reconstructs defect features by calculating time delays and multiplying signals to enhance spatial coherence and suppress incoherent noise and artefacts. Experiments show that DE-DMAS significantly improves resolution for shallow surface defects. Compared to the Total Focusing Method (TFM), DE-DMAS improves the Contrast Ratio (CR) and Contrast-to-Noise Ratio (CNR) by up to 98% and 184.6%, respectively. This method offers a new approach for detecting shallow surface defects in rails.
Zhang et al. (Wed,) studied this question.