Exploiting the Capabilities of Surfcam for Coastal Morphodynamic Analysis

• Published in: Journal of coastal research Vol. 95; no. sp1; pp. 1333 - 1338
• Main Authors: Valentini, Nico, Balouin, Yann, Bouvier, Clement
• Format: Journal Article
• Language: English
• Published: Fort Lauderdale Coastal Education and Research Foundation 26.05.2020; Allen Press Publishing; Allen Press Inc
• Series: SI95
• Subjects: Algorithms; Analysis; Bathymeters; Bathymetry; bathymetry estimation; Beaches; CNN; Coastal research; depth-inversion; Earth Sciences; Exploitation; Geomorphology; Gravel; Image processing; Image reconstruction; Image segmentation; Intertidal environment; Intertidal zone; METHODOLOGIES; Mountain regions; Mountainous areas; Neural networks; Sciences of the Universe; Shorelines; Surfcam; Tidal range; Wave energy; Wave power; CNN; Depth-inversion; ADDITIONAL INDEX WORDS: Surfcam; Image-processing; Bathymetry estimation
• ISSN: 0749-0208; 1551-5036
• DOI: 10.2112/SI95-256.1

Valentini, N.; Balouin, Y., and Bouvier, C., 2020. Exploiting the capabilities of Surfcam for coastal morphodynamic analysis. In: Malvárez, G. and Navas, F. (eds.), Global Coastal Issues of 2020. Journal of Coastal Research, Special Issue No. 95, pp. 1333–1338. Seville (Spain), ISSN 0749-0208. A free webcam network is used in this study, operated by ©Viewsurf, a European company with more than 1200 stations in France, among beaches, urban and mountain areas in order to assess the capabilities of surfcam for coastal morphodynamics analysis. Four representative coastal sites, with different tidal range and wave energy are monitored in order to extract optical signatures for morphodynamic analysis: three in the Atlantic and a gravel beach in the Mediterranean area. A representative site is here documented, Hendaye beach, which is characterized by a meso-tidal environment, with highly energetic waves often present, which makes the image processing challenging. Shoreline detections for beach width and intertidal area calculation are accomplished by using a new image segmentation algorithm, based on a framework, derived by opportunely coupling a superpixels algorithm and a Deep-Conditional-Neural-Network DCNN model. Aiming to exploit the capabilities of cBathy algorithm (Holman, Plant, and Holland, 2013), a version of the open-source methodology (https://coastal-imaging-research-network.github.io), is employed for bathymetry reconstruction. A coupled bathymetry by merging cBathy depth estimate and intertidal reconstructed depth from automatically extracted shorelines is accomplished, which is compared with survey and lead to an RMS error of 0.43 m. Discussions are given with respect to this new potential open-source resource and its limitations.