Image denoising of ultrasonic echo data acquired on concrete
- Ultrasonic echo testing has become a common method in civil engineering for the investigation of concrete structures. The detection of inhomogeneities, reinforcing elements and the geometry of the object is required for quality assurance and Inspection. This assessment depends on the quality of ultrasonic images which can be improved by using Reverse Time Migration (RTM) rather than the standard method, Synthetic Aperture Focusing Technique (SAFT). Although RTM provides a better mapping of circular objects and (dipping) reflectors, the image is corrupted by migration noise. To suppress the image noise, we have tested various filter methods in the spatial domain, frequency domain as well as in the curvelet domain on ultrasonic RTM images. We found that either a spatial edge detection filter in combination with a lowpass filter (Laplacian of Gaussian filter) or two lowpass filter with different filter parameters (Difference of Gaussian filter) removed artefacts. An additional smoothingUltrasonic echo testing has become a common method in civil engineering for the investigation of concrete structures. The detection of inhomogeneities, reinforcing elements and the geometry of the object is required for quality assurance and Inspection. This assessment depends on the quality of ultrasonic images which can be improved by using Reverse Time Migration (RTM) rather than the standard method, Synthetic Aperture Focusing Technique (SAFT). Although RTM provides a better mapping of circular objects and (dipping) reflectors, the image is corrupted by migration noise. To suppress the image noise, we have tested various filter methods in the spatial domain, frequency domain as well as in the curvelet domain on ultrasonic RTM images. We found that either a spatial edge detection filter in combination with a lowpass filter (Laplacian of Gaussian filter) or two lowpass filter with different filter parameters (Difference of Gaussian filter) removed artefacts. An additional smoothing was obtained by applying the first generation curvelet transform after downsampling the image matrix and adding Gaussian noise. The proposed filter scheme is able to suppress RTM noise and enhance the image quality such that the objective interpretation of ultrasonic images for Quality assessment of concrete specimen is simplified.…
Autor*innen: | Sarah Sieber |
---|---|
Weitere Person: | Ernst NiederleithingerORCiD, Maria Grohmann |
Dokumenttyp: | Posterpräsentation |
Veröffentlichungsform: | Präsentation |
Sprache: | Englisch |
Jahr der Erstveröffentlichung: | 2016 |
DDC-Klassifikation: | Technik, Medizin, angewandte Wissenschaften / Ingenieurwissenschaften / Ingenieurwissenschaften und zugeordnete Tätigkeiten |
Freie Schlagwörter: | Curvelet Transform; Image Denoising; Reverse-time migration; Ultrasonic echo technique |
Veranstaltung: | 76. Jahrestagung der Deutschen Geophysikalischen Gesellschaft |
Veranstaltungsort: | Münster, Germany |
Beginndatum der Veranstaltung: | 14.03.2016 |
Enddatum der Veranstaltung: | 17.03.2016 |
Verfügbarkeit des Dokuments: | Datei im Netzwerk der BAM verfügbar ("Closed Access") |
Datum der Freischaltung: | 28.04.2016 |
Referierte Publikation: | Nein |