Ultrasonic imaging method and system for detecting delamination defects in composite aircraft skin
By extracting grayscale, texture, and frequency domain features from the ultrasonic inspection of composite aircraft skin, and combining them with a binary classifier, the problem of distinguishing between layered defects and normal areas was solved, achieving efficient and reliable layered defect detection and reducing the false positive rate and missed detection rate.
Patent Information
- Application Number
- CN202610664561.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-05-14
- Publication Date
- 2026-07-17
AI Technical Summary
Existing technologies struggle to effectively distinguish delamination defects from normal resin-rich areas in ultrasonic testing of composite aircraft skins, resulting in a high rate of missed detection for both micro-delamination and deep, weak-signal delamination defects. Furthermore, deep learning methods that rely on large-scale labeled samples lack physical interpretability.
By extracting grayscale statistics, texture, and frequency domain features from ultrasonic C-scan grayscale images, and combining them with a binary classifier for defect identification and quantitative analysis, the differences in the weaving period peaks of composite materials in the spectrum are used to distinguish between layered defects and normal areas. Adaptive nonlocal mean filtering for noise reduction and multi-feature fusion are also employed.
It achieves a high detection rate (97%) and a low false positive rate (3%) for layered defects with an area greater than 5 mm², without the need for large-scale labeled sample training. It has good physical interpretability and engineering applicability, and reduces the complexity of detection and the misjudgment rate.
Smart Images
Figure CN122409875A_ABST
Abstract
Citation Information
Patent Citations
Carbon fiber composite material defect damage classification identification and quantitative analysis method and system
CN113012123A