Image Algorithm of Ship Detection for Surface Vehicle

In order to solve the problem that the surface vehicle detecting ship targets in the nearshore waters was vulnerable to light, similar color background and wave reflection, ship detection algorithm based on improved visual attention model was proposed. First, low frequency and high frequency features of images were extracted by using wavelet transform theory. Then, the hue, saturation and value features of images were also extracted by converting the images of task waters from RGB color space to HSV color space. Finally, various features of images were merged in the application of image processing method of Gaussian Pyramid and normalization operator. The simulation results show that the proposed ship detection method can accurately detect the ship targets under complicated backgrounds and has satisfactory anti-interference capability.