Quantitative Detection of Near-Surface Defects in Polymer Materials Based on Focused Ultrasound Thermal Effect

The detection and quantitative evaluation of internal defects in polymer materials are critical for practical engineering applications. This paper introduces a method for the quantitative evaluation of defects in polymer materials using the ultra-sound thermal effect. Initially, a heat generation model for polymer materials under ultrasonic excitation is developed, and based on heat conduction theory, the mechanism by which defect characteristics affect the temperature distribution in infrared images is examined. A hybrid simulation and experimental approach is employed to analyze the variation patterns of infrared images on the surfaces of materials with various defects under the ultrasound thermal effect. A method is proposed to accurately assess the size and location of defects by measuring the width of the opening and the deepest concave point in the anomalous regions of infrared images. Local image frequency variations are assessed through normalized local gradients and variance, while an adaptive fractional-order model is developed to enhance the images. Morphological operations are employed to reconstruct grayscale images, with the resulting images serving as foreground markers for subsequent segmentation using the watershed algorithm. The convex hull algorithm is utilized to compute the width of the opening and the location of the deepest concave point in the infrared images, thereby facilitating quantitative defect analysis. This approach offers a novel perspective for the quantitative evaluation of internal defects in polymer materials.