Edge detection and depth evaluation of CFRP internal delamination defects based on differential image frequency domain preprocessing

Abstract Carbon fiber reinforced polymers (CFRP) are widely used in fields such as aviation and aerospace. However, subtle defects can significantly impact the material's service life, making defect detection a critical priority. This paper presents a method for detecting delamination defects in carbon fiber reinforced composites (CFRP) using line laser infrared thermography. A preprocessing approach combining differential thermography and frequency-domain filtering is proposed to effectively eliminate the trailing artifacts caused by line laser scanning, resulting in defect feature images with an improved signal-to-noise ratio. Utilizing the preprocessed frequency-domain magnitude image data, an improved Fuzzy C-Means clustering segmentation algorithm is developed, achieving high-precision edge detection of layered defects with an average accuracy of 95.0%. Furthermore, defect depth classification based on the frequency-domain magnitude data is performed using the K-Nearest Neighbor (KNN) algorithm, yielding an average accuracy of 98.8%. These results validate the effectiveness of the proposed algorithms for defect edge detection and depth assessment.