Perceptual Fusion of Electronic Chart and Marine Radar Image

Electronic charts and marine radars are indispensable equipment in ship navigation systems, the fusion display of these two parts ensures that vessel can dangerous moving targets various obstacles on sea. To reduce noise interference caused by external factors hardware, a novel radar image denoising algorithm using concept Generative Adversarial Network (GAN) Wasserstein distance is proposed. GAN focuses transferring distribution between strong weak noise, while perceptual loss approach to suppress comparing characteristics output after denoising. Afterwards, an registration method based transformation proposed eliminate imaging difference chart image, which visual attribute transfer used transform images. Finally, sparse theory process high frequency low subband coefficients detection obtained fast Fourier parallel realizing fusion. The results show fused contour has consistency, training speed short time.