Estimating wave height based on feature fusion of ship motion response and X-band radar images

Since estimating the encountered sea wave is particularly important, retrieving significant wave height (SWH) from ship motion response (SMR) and X-band marine radar images is investigated. Although SMR and radar images have been widely utilized separately for estimating SWH, the retrieving accuracy and reliability under complex nonlinear sea conditions should be further enhanced. In this paper, the features of SMR and the radar images are investigated. A feature fusion method based on the statistical characteristic extracted from SMR historical data and the signal-to-noise ratio (SNR) feature achieved from radar images is proposed for estimating SWH. The feature vector and observation matrix are constructed using the significant values extracted from SMR historical data and the wave SNR extracted from radar images. Then, the weight coefficient of the feature vector, which is used to estimate SWH, is determined based on the least-squares fitting algorithm. Using the simulated SMR historical data and radar images, the experimental results demonstrate that the proposed feature fusion approach has better retrieval accuracy than the SNR-based approach; the colorization coefficient (CC) of the retrieved SWH approximately approaches 1, and the root mean square error (RMSE) decreases to 0.14 m.