Journals Articles Conferences News About CLP
Submit a paper Email Alert
Search
Search
Articles
Journals
News
Advanced Search
Top Searches
2D MaterialsTransformation opticsQuantum PhotonicsSmall lasersSemiconductor UV PhotonicsSupersymmetric microring laser arrays

Laser & Optoelectronics Progress, Volume. 58, Issue 24, 2401001(2021)

Nearshore Wave Period Detection Based on Video Spatiotemporal Feature Learning

Wei Song*, Yuanyuan Chen, Qi He, and Yanling Du**
Author Affiliations
  • College of Information Technology, Shanghai Ocean University, Shanghai 201306, China
    • show less
    AbstractGet PDF(in Chinese)
    Figures&Tables (8)
    Equations (0)
    References (39)
    Tools
    Paper Information
    Topics
    Figures & Tables(8)
    Framework of wave period detection model

    Fig. 1. Framework of wave period detection model

    Download full size
    Wave period detection model based on spatiotemporal fusion composite convolutional neural network

    Fig. 2. Wave period detection model based on spatiotemporal fusion composite convolutional neural network

    Download full size
    ConvLSTM model for wave period detection

    Fig. 3. ConvLSTM model for wave period detection

    Download full size
    Prediction curves of CNN-2D1D model and C3D model on August 22, 2018 and August 23, 2018. (a) 2018-08-22; (b) 2018-08-23

    Fig. 4. Prediction curves of CNN-2D1D model and C3D model on August 22, 2018 and August 23, 2018. (a) 2018-08-22; (b) 2018-08-23

    Download full size

    • Table 1. Distribution and construction of datasets

      View table

      Table 1. Distribution and construction of datasets

      DatasetNumber of samples (training set/validation set/test set)Size of single sampleWave period /s
      Dataset Ⅰ29400/4900/490064×64×34.3--12.7
      Dataset Ⅱ28959/4459/445930×64×64×3
      Dataset Ⅲ27979/3479/347910×64×64×3

    • Table 2. Wave period detection results of different models on test set

      View table

      Table 2. Wave period detection results of different models on test set

      ModelDatasetMAERMSE
      VGG16Dataset Ⅰ0.931.41
      ConvLSTM10Dataset Ⅱ1.081.57
      CNN2D1D-10Dataset Ⅱ1.211.66
      ConvLSTM30Dataset Ⅲ0.821.13
      CNN2D1D-30Dataset Ⅲ0.480.66
      C3DDataset Ⅲ0.470.81

    • Table 3. Detection errors of C3D and CNN-2D1D models on August 22, 2018 and August 23, 2018

      View table

      Table 3. Detection errors of C3D and CNN-2D1D models on August 22, 2018 and August 23, 2018

      ErrorCNN-2D1DC3D
      August 22, 2018August 23, 2018August 22, 2018August 23, 2018
      MAE0.180.990.461.01
      RMSE0.261.210.531.31

    • Table 4. Comparison of model sizes of CNN-2D1D and C3D models

      View table

      Table 4. Comparison of model sizes of CNN-2D1D and C3D models

      ModelCNN-2D1DC3D
      Parameters15, 019, 20252, 830, 593
      Model size /MB173605
    Tools

    Get Citation

    Copy Citation Text

    Wei Song, Yuanyuan Chen, Qi He, Yanling Du. Nearshore Wave Period Detection Based on Video Spatiotemporal Feature Learning[J]. Laser & Optoelectronics Progress, 2021, 58(24): 2401001

    Download Citation

    EndNote(RIS)BibTexPlain Text
    Set citation alerts for article
    Save article for my favorites
    Paper Information

    Category: Atmospheric Optics and Oceanic Optics

    Received: Nov. 30, 2020

    Accepted: Feb. 17, 2021

    Published Online: Nov. 24, 2021

    The Author Email: Wei Song (wsong@shou.edu.cn), Yanling Du (yldu@shou.edu.cn)

    DOI:10.3788/LOP202158.2401001

    Topics

    laser devices and laser physics
    Lasers and Laser Optics
    Laser physics
    laser manufacturing
    Instrumentation, Measurement and Metrology