Wave Direction Estimation from Optical Satellite Imagery Using Multi-Stage Gabor Filter

The direction of incoming waves is a primary force driving coastal erosion, dictating the sediment transport patterns that shape shorelines. Accurate estimation of this parameter is therefore essential for effective coastal management. This paper proposes a robust method for estimating wave direction from high-resolution satellite image patches. The approach utilizes texture analysis based on Gabor filters to identify dominant wave crest orientations, with preprocessing enhanced by contrast limited adaptive histogram equalization (CLAHE). We evaluate three orientation strategies: a baseline standard detection, a high-precision two-stage refinement, and an interpolation-based refinement. The methodology is first validated on a synthetic benchmark image with a known ground truth to establish baseline performance, and then applied to real-world coastal imagery from Thailand. Experiments show that refinement strategies, particularly the two-stage method, significantly improve accuracy in ideal conditions and deliver stable, interpretable estimates on real-world data. On the synthetic benchmark, the two-stage refinement achieved a low mean absolute error (MAE) of 0.234°, while on real-world coastal imagery, the interpolation refinement demonstrated strong performance with an MAE of 12.95°. Ultimately, this method provides a scalable, non-intrusive tool for analyzing nearshore wave behavior, while emphasizing that optimal results depend on consistent preprocessing and parameter tuning.

• Publication year

  2025

• Venue

  Information and Communication Technology Convergence

• Publication date

  2025-10-14

• Fields of study

  Not labeled

• Identifiers
  DOI 10.1109/ICTC66702.2025.11388568
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Dual-Modal Approach for Ship Detection: Fusing Synthetic Aperture Radar and Optical Satellite Imagery

  2025cited by this paper
• Geoinformatics and Machine Learning for Shoreline Change Monitoring: A 35-Year Analysis of Coastal Erosion in the Upper Gulf of Thailand

  2025cited by this paper
• Advances in Remote Sensing and Deep Learning in Coastal Boundary Extraction for Erosion Monitoring

  2025cited by this paper
• The future of coastal monitoring through satellite remote sensing

  2022cited by this paper
• Multiparametric sea state fields from synthetic aperture radar for maritime situational awareness

  2022cited by this paper
• Sea State from Single Optical Images: A Methodology to Derive Wind-Generated Ocean Waves from Cameras, Drones and Satellites

  2021cited by this paper
• Investigation of edge detection methods using the example of wave field pattern image

  2021cited by this paper
• The Analytical of Wave Characteristics and Shoreline Changes at Tumbon Kao Roop Chang, Songkhla Province in 2015 and 2016

  2017cited by this paper
• Remote sensing of the nearshore.

  2013cited by this paper
• Evaluation of the effects of Gabor filter parameters on texture classification

  2007cited by this paper
• Wave Modelling - The State of the Art

  2007cited by this paper
• Extraction of coastal ocean wave fields from SAR images

  2005cited by this paper
• Directional analysis of SAR images aiming at wind direction

  2004cited by this paper
• Comparison and fusion of co‐occurrence, Gabor and MRF texture features for classification of SAR sea‐ice imagery

  2001cited by this paper
• Coastal Geomorphology: An Introduction

  2001cited by this paper
• Wave directional spectrum from SAR imagery

  2000cited by this paper
• Longshore sand transport on beaches

  1970cited by this paper
• Evaluation of the effects of Gabor filter parameters on texture classification

  year unknowninfluential reference

• No citing papers are available for this paper.