Beach State Recognition Using Argus Imagery and Convolutional Neural Networks

Abstract

Nearshore morphology is a key driver in wave breaking and the resulting nearshore circulation, recreational safety, and nutrient dispersion. Morphology persists within the nearshore in specific shapes that can be classified into equilibrium states. Equilibrium states convey qualitative information about bathymetry and relevant physical processes. While nearshore bathymetry is a challenge to collect, much information about the underlying bathymetry can be gained from remote sensing of the surfzone. This study presents a new method to automatically classify beach state from Argus daytimexposure imagery using a machine learning technique called convolutional neural networks (CNNs). The CNN processed imagery from two locations: Narrabeen, New South Wales, Australia and Duck, North Carolina, USA. Three different CNN models are examined, one trained at Narrabeen, one at Duck, and one trained at both locations. Each model was tested at the location where it was trained in a self-test, and the single-beach models were tested at the location where it was not trained in a transfer-test. For the self-tests, skill (as measured by the F-score) was comparable to expert agreement (CNN F-values at Duck = 0.80 and Narrabeen = 0.59). For the transfer-tests, the CNN model skill was reduced by 24–48%, suggesting the algorithm requires additional local data to improve transferability performance. Transferability tests showed that comparable F-scores (within 10%) to the self-trained cases can be achieved at both locations when at least 25% of the training data is from each site. This suggests that if applied to additional locations, a CNN model trained at one location may be skillful at new sites with limited new imagery data needed. Finally, a CNN visualization technique (Guided-Grad-CAM) confirmed that the CNN determined classifications using image regions (e.g., incised rip channels, terraces) that were consistent with beach state labelling rules.

RETRACTED: PSCNN: PatchShuffle Convolutional Neural Network for COVID-19 Explainable Diagnosis

“…At last, gradient-weighted class activation mapping (Grad-CAM) (35) is harnessed to deliver explanations on how our PSCNN model creates the decision. The output of nCM-5 in Figure 9 is chosen for Grad-CAM.…”

Coastal Image Classification and Pattern Recognition: Tairua Beach, New Zealand

“…Ellenson et al [39] used classical CNN models (ResNet50) to classify beach states into five classes according to Wright and Short [2]'s classification scheme. They conducted experiments to evaluate the performance of their CNN model based on imagery datasets from Narrabeen (New South Wales, Australia) and Duck (North Carolina, USA) and showed that CNN had accurately identified key features of the coastal images which distinguished beach states.…”

“…The best result in each category is marked in bold. We analyzed the F1-score for the seven beach states using two different classification networks: ResNet50 (used by Ellenson et al [39]) and ResNext50 (as seen in Table 3). In the F1-score of the two networks, the score of type A and type G is obviously higher than the others.…”

Labeling Poststorm Coastal Imagery for Machine Learning: Measurement of Interrater Agreement

“…In some situations, labels can be assigned automatically-e.g., merging time-stamped images with time-stamped sensor data (e.g., Buscombe & Carini, 2019;Buscombe et al, 2020). But most of the time, labeling cannot be done programmatically and instead requires human interpretation (e.g., Ellenson et al, 2020;Liu et al, 2014;Buscombe & Ritchie, 2018;Morgan et al, 2019;Yang et al, 2021).…”