Multispectral fusion-based crack detection using automated laser thermography inspection
- Crack detection of complex components is critical for maintaining the integrity and safety of structures across industries such as energy production and aerospace. The integration of multispectral imaging and automatic robot-based non-destructive testing (NDT) marks a significant advancement in the inspection of complex geometries. Multispectral imaging leverages the utilization of image data from different wavelength ranges for enhanced feature extraction and analysis. This paper proposes a novel image fusion model by combining thermal and visual images for the detection of surface and sub-surface cracks of turbine blades in energy production. Our study uses 3D scanning of components using a robot-assisted flying laser line, which measures high-resolution thermal data and RGB images using visual cameras simultaneously. Laser thermography induces a localized heating on the sample, creating a thermal response that highlights cracks and defects that are often undetectable usingCrack detection of complex components is critical for maintaining the integrity and safety of structures across industries such as energy production and aerospace. The integration of multispectral imaging and automatic robot-based non-destructive testing (NDT) marks a significant advancement in the inspection of complex geometries. Multispectral imaging leverages the utilization of image data from different wavelength ranges for enhanced feature extraction and analysis. This paper proposes a novel image fusion model by combining thermal and visual images for the detection of surface and sub-surface cracks of turbine blades in energy production. Our study uses 3D scanning of components using a robot-assisted flying laser line, which measures high-resolution thermal data and RGB images using visual cameras simultaneously. Laser thermography induces a localized heating on the sample, creating a thermal response that highlights cracks and defects that are often undetectable using conventional imaging techniques. Multispectral image fusion models are developed based on an unsupervised generative adversarial network (GAN) for precise crack detection. The model validation is done using experimental data, and it demonstrated high crack detection accuracy compared to conventional single-band imaging methods. This research highlights the new potential for advancing NDT technologies by combining thermal inspection techniques with optical imaging to develop a multispectral inspection approach.…


| Autor*innen: | Sruthi Krishna Kunji PurayilORCiD |
|---|---|
| Koautor*innen: | Philipp Daniel HirschORCiD, Julien Lecompagnon, Mathias Ziegler |
| Dokumenttyp: | Vortrag |
| Veröffentlichungsform: | Präsentation |
| Sprache: | Englisch |
| Jahr der Erstveröffentlichung: | 2025 |
| Organisationseinheit der BAM: | 8 Zerstörungsfreie Prüfung |
| 8 Zerstörungsfreie Prüfung / 8.3 Thermografische Verfahren | |
| DDC-Klassifikation: | Technik, Medizin, angewandte Wissenschaften / Ingenieurwissenschaften / Ingenieurbau |
| Freie Schlagwörter: | Infrared imaging; Laser; Multispectral imaging; NDE 4.0; Non-destructive testing |
| Themenfelder/Aktivitätsfelder der BAM: | Infrastruktur |
| Infrastruktur / Leichtbau | |
| Veranstaltung: | SPIE Defense+Commercial Sensing- Thermosense: Thermal Infrared Applications XLVII |
| Veranstaltungsort: | Orlando, FL, USA |
| Beginndatum der Veranstaltung: | 13.04.2025 |
| Enddatum der Veranstaltung: | 17.04.2025 |
| Verfügbarkeit des Dokuments: | Datei im Netzwerk der BAM verfügbar ("Closed Access") |
| Datum der Freischaltung: | 06.05.2025 |
| Referierte Publikation: | Nein |
| Eingeladener Vortrag (wissenschaftliche Konferenzen): | Nein |

