skip to main content

Explore Research Products in the NSF-PAR

Title: A Data-Driven Approach to Classifying Wave Breaking in Infrared Imagery
We apply deep convolutional neural networks (CNNs) to estimate wave breaking type (e.g., non-breaking, spilling, plunging) from close-range monochrome infrared imagery of the surf zone. Image features are extracted using six popular CNN architectures developed for generic image feature extraction. Logistic regression on these features is then used to classify breaker type. The six CNN-based models are compared without and with augmentation, a process that creates larger training datasets using random image transformations. The simplest model performs optimally, achieving average classification accuracies of 89% and 93%, without and with image augmentation respectively. Without augmentation, average classification accuracies vary substantially with CNN model. With augmentation, sensitivity to model choice is minimized. A class activation analysis reveals the relative importance of image features to a given classification. During its passage, the front face and crest of a spilling breaker are more important than the back face. For a plunging breaker, the crest and back face of the wave are most important, which suggests that CNN-based models utilize the distinctive ‘streak’ temperature patterns observed on the back face of plunging breakers for classification.  more » « less
Award ID(s):
1736389
NSF-PAR ID:
10201659
Author(s) / Creator(s):
;
Date Published:
Journal Name:
Remote Sensing
Volume:
11
Issue:
7
ISSN:
2072-4292
Page Range / eLocation ID:
859
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. ; ; ( , Journal of Geophysical Research: Oceans)
    Abstract

    This is the second of a two‐part series concerning remote observation and wave‐by‐wave analysis of the onset of breaking for spilling and plunging waves in the surf zone. Nearshore phase‐averaged and phase‐resolving wave models parameterize and directly simulate wave breaking and require realistic critical values of key wave parameters, such as the depth‐limited breaking indexγ, steepness, or phase speed to initialize wave breaking. Using LIDAR line‐scans and infrared imagery, we observe over 1,600 breaking waves at the US Army Corps of Engineers Field Research Facility (FRF) in Duck, NC, and examine these parameters on a wave‐by‐wave basis at the onset of breaking for 413 spilling and 111 plunging waves. We find thatγis maximum near the onset of breaking at values consistent with those previously observed at the FRF, but thatγfor plunging waves (0.73 ≤ γP ≤ 0.81) is greater thanγfor spilling waves (0.63 ≤ γS ≤ 0.71). Direct estimates of wave face slope are maximum at the onset of breaking, approximately 22° for spilling and 30° for plunging waves. Using the relationship betweenγand wave face slope, we develop a threshold for the onset of breaking that is a linear function of the two parameters. Wave face slope andγare further used together to quantify whether a spilling‐ or plunging‐type breaker is more likely. We test the Miche steepness limit on our depth‐limited breaking data and find it correctly predicts only 10% of the plunging breakers and none of the spilling breakers in the surf zone.

     
    more » « less
  2. ; ; ( , Journal of Geophysical Research: Oceans)
    Abstract

    This is the first of a 2‐part series concerning remote observation and wave‐by‐wave analysis of the onset of breaking in the surf zone. In the surf zone, breaking waves drive nearshore circulation, suspend sediment, and promote air‐sea gas exchange. Nearshore wave model predictions often diverge fromin situmeasurements near the break point location because common parameterizations do not account for the rapid changes that occur near the onset of breaking. This work presents extensive methodology to combine data from a line‐scanning LIDAR and thermal infrared cameras to detect breaking, classify breaker type, and measure geometric wave parameters on a wave‐by‐wave basis, which can be used to improve breaker parameterizations. Over 2,600 non‐breaking and 1,600 breaking waves are analyzed from data collected at the USACE Field Research Facility in Duck, NC, including 413 spilling and 111 plunging waves for which the onset of breaking was observed. Wave height is estimated using a spatio‐temporal method for wave tracking that preserves the sea surface elevation maximum and overcomes field of view limitations. Methods for estimating instantaneous wave speed are refined by fitting a skewed Gaussian function to each wave profile before tracking the peaks. Wave slope is estimated from a linear fit to the upper 80% of the wave face, which provides a robust metric and strong correlation with geometric wave slope defined relative to mean sea level. Finally, breaking wave face foam coverage is analyzed to assess common model assumptions about roller length for wave energy dissipation parameterizations.

     
    more » « less
  3. ; ; ; ; ( , Journal of Fluid Mechanics)

    An experimental study of the dynamics and droplet production in three mechanically generated plunging breaking waves is presented in this two-part paper. In the present paper (Part 2), in-line cinematic holography is used to measure the positions, diameters ($d\geq 100\ \mathrm {\mu }{\rm m}$), times and velocities of droplets generated by the three plunging breaking waves studied in Part 1 (Erininet al.,J. Fluid Mech., vol. 967, 2023, A35) as the droplets move up across a horizontal measurement plane located just above the wave crests. It is found that there are four major mechanisms for droplet production: closure of the indentation between the top surface of the plunging jet and the splash that it creates, the bursting of large bubbles that were entrapped under the plunging jet at impact, splashing and bubble bursting in the turbulent zone on the front face of the wave and the bursting of small bubbles that reach the water surface at the crest of the non-breaking wave following the breaker. The droplet diameter distributions for the entire droplet set for each breaker are fitted with power-law functions in separate small- and large-diameter regions. The droplet diameter where these power-law functions cross increases monotonically from 820 to 1480$\mathrm {\mu }{\rm m}$from the weak to the strong breaker, respectively. The droplet diameter and velocity characteristics and the number of the droplets generated by the four mechanisms are found to vary significantly and the processes that create these differences are discussed.

     
    more » « less
  4. ; ; ; ; ; ( , Geophysical Research Letters)
    Abstract

    An experimental investigation of droplet generation by a plunging breaking wave is presented. In this work, simultaneous measurements of the wave crest profile evolution and of droplets ranging in radius down to 50 μm for a mechanically generated plunging breaker during many repeated breaking events in freshwater are performed. We find three distinct time zones of droplet production, first when the jet impacts the free surface upstream of the wave crest, second when the large air bubbles entrapped by the plunging jet impact reach the free surface and burst, and third when smaller bubbles burst upon reaching the free surface later in the breaking process. These subprocesses account for 22%, 44%, and 34%, respectively, of the average of 653 droplets produced per breaking event. The probability distributions of the ranges of large and small droplet radii are well represented by power law functions that intersect at a radius of 418 μm.

     
    more » « less
  5. ; ; ; ( , Journal of Fluid Mechanics)

    An experimental study of the dynamics and droplet production in three mechanically generated plunging breaking waves is presented in this two-part paper. In the present paper (Part 1), the dynamics of the three breakers are studied through measurements of the evolution of their free surface profiles during 10 repeated breaking events for each wave. The waves are created from dispersively focused wave packets generated with three wave maker motions that differ primarily by small changes in their overall amplitude. Breaker profiles are measured with a cinematic laser-induced fluorescence technique covering a streamwise region of approximately one breaker wavelength and over a time of 3.4 breaker periods. The aligned profile data is used to create spatio-temporal maps of the ensemble average surface height and the standard deviation of both the local normal distance and the local arc length relative to the instantaneous mean profile. It is found that the mean and standard deviation maps contain strongly correlated localized features and indicate that the transition from laminar to turbulent flow is a highly repeatable process. Regions of high standard deviation include the splash created by the plunging jet impact and subsequent splash impacts at the front of the breaking region as well as the site where the air pocket entrained under the plunging jet at the moment of jet tip impact comes to the surface and pops. In Part 2 (Erininet al., J. Fluid Mech., vol. 967, 2023, A36), these features are used to interpret various features of the distributions of droplet number, diameter and velocity.

     
    more » « less
  • Citation Formats
  • MLA
  • APA
  • Chicago
  • BibTeX
  • Save / Share this Record
  • Send to Email