Physics-Guided Deep Learning Framework Estimates Coastal Wave Parameters from Video
Researchers developed a physics-informed deep learning system that estimates nearshore wave peak periods directly from coastal video footage, combining transformer and recurrent-convolutional architectures with physics-based regularization. The approach addresses limitations of traditional monitoring systems like buoys and radar by offering lower-cost, broader spatial coverage alternatives for coastal engineering and hazard assessment. The framework demonstrates potential for long-term, operationally feasible coastal monitoring critical for climate resilience and shoreline protection.
A new physics-guided spatiotemporal learning framework uses passive video monitoring to estimate nearshore wave peak periods, a key parameter for coastal engineering and marine hazard assessment. The system combines automated region-of-interest detection, multi-stage transfer learning from synthetic to real data, and physics-informed regularization to improve accuracy and physical consistency. Testing multiple architectures showed transformer-based models achieved higher instantaneous prediction accuracy, while lightweight recurrent-convolutional models provided better temporal stability and operational oceanographic skill. Ablation studies confirmed that physics-guided regularization reduced physically implausible predictions and improved trend-following consistency. Explainability analysis revealed the model focused appropriately on hydrodynamically active surf-zone regions, aligning with known wave propagation physics.
What's missing
The paper does not specify the geographic location(s) where the video data was collected, the temporal duration of the dataset, or quantitative performance metrics (e.g., mean absolute error, correlation coefficients) comparing the proposed framework directly against traditional buoy or radar measurements in the same locations.
What different sources said
- arXiv cs.AICenter
Physics-Guided Spatiotemporal Learning for Coastal Wave Peak Period Estimation from Video
Related
Topology-Aware Thermodynamics Improves DNA Probe Specificity Design
Researchers developed a new framework for designing DNA probes that accounts for the spatial organization of matched sequences, not just overall thermodynamic stability. Traditional methods rely on scalar measures like melting temperature and free energy, which miss how mismatches are distributed along the probe. The approach could improve diagnostic accuracy in applications like HPV detection and gene expression profiling.
Study Identifies Optimal Thermal Dose for Combining Focused Ultrasound with Immunotherapy in Tumors
Researchers used multimodal PET imaging to identify an optimal thermal dose range for focused ultrasound ablation that destroys tumor tissue while preserving conditions for immunotherapy delivery. The study found that excessive heating collapses blood vessels needed for antibody access, while insufficient heating fails to adequately reduce tumor burden. The findings could guide clinical design of combination treatments pairing thermal ablation with immunotherapies.
Plant MSH1 Protein Functions as Mismatch-Directed Nuclease for Organelle Genome Maintenance
Researchers have identified the precise mechanism by which the AtMSH1 protein in Arabidopsis plants recognizes and cleaves DNA mismatches and lesions, preventing mutations in organellar genomes. The protein combines a DNA mismatch recognition module with a nuclease domain that makes staggered cuts at specific positions relative to DNA damage. This discovery explains how plants maintain unusually low mutation rates in their mitochondrial and chloroplast DNA compared to other eukaryotes.