CNN Wave Radar Inversion for Nonlinear Sea-State Reconstruction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for inferring sea-state parameters from X-band radar images are limited by shadowing modulation, leading to poor reconstruction of nonlinear wave components, especially in deep water environments and aboard ships.
Innovation Solution
A method using a convolutional neural network (CNN) autoencoder is developed to invert X-band radar images, utilizing a training dataset of simulated radar images and corresponding ocean surface wave height maps, with a specific encoder-decoder structure and supervised training to map radar images to wave height maps, capable of handling noise and obstructions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional mathematical and physical methods are used to invert X-band radar images, then the inversion process can be performed with simpler computational requirements, but the reconstruction of nonlinear wave components suffers from poor performance due to shadowing modulation
Solution Approach 1:
The patent replaces traditional mathematical and physical inversion methods with a deep learning-based computational model. The CNN automatically learns the complex mapping between radar images and wave parameters, substituting the need for explicit physical models and empirical modulation transfer functions. This enables accurate reconstruction of nonlinear wave components including those affected by shadowing modulation, while maintaining computational efficiency through the trained neural network.
2Adaptability or versatility
If empirical modulation transfer functions are used to handle shadowing modulation, then the inversion can proceed with available data, but the performance deteriorates for particular sea states with poor reconstruction of nonlinear wave components
Solution Approach 1:
The patent transforms the inversion approach by changing from fixed empirical parameters to adaptive learned parameters. The deep learning model is trained on diverse sea state conditions, enabling it to adapt to different wave conditions including those with shadowing modulation. The model learns optimal parameter transformations during training, achieving both broad adaptability across sea states and high precision in reconstructuring nonlinear wave components.
3Measurement precision
If more complex traditional inversion methods are developed to improve nonlinear component reconstruction, then measurement precision may improve, but the device complexity and computational requirements increase significantly
Solution Approach 1:
The patent substitutes complex traditional inversion algorithms with a deep learning-based computational model. The CNN automatically learns the complex mapping between radar images and wave parameters, replacing the need for complicated mathematical transformations and empirical corrections. This achieves high-precision reconstruction of nonlinear wave components while maintaining computational efficiency through the trained neural network's ability to perform inversions rapidly once trained.
Data Source
AI summary
A method of convolutional neural network (CNN) based inversion of wave radar images and use thereof pertain to the field of sea-state monitoring, in which a CNN autoencoder is employed. A training dataset produced by a simulation tool is input to the CNN autoencoder for supervised training, with X-band radar images of an inversion region being taken as input to the training dataset and with corresponding ocean surface wave height maps being taken as output from the training dataset, thereby deriving a model defining a mapping of them. The CNN autoencoder includes an encoder, fully connected layers and a decoder. The encoder includes five convolutional layers and five max-pooling layers, which are connected alternately. There are two symmetric fully connected layers, and the decoder includes five deconvolutional layers. A CNN-based deep learning algorithm is used to provide good reconstruction of the nonlinear components. The autoencoder technology is employed, which can directly reconstruct real-time ocean surface wave height maps from X-band radar wave images with minor errors, which provide more comprehensive sea-state information.