Sea Sky Line Detection Based on Edge Phase Encoding in Complicated Background

Abstract

The complex sea sky background such as clouds, wave reflection, complex weather, will bring difficulties to the sea sky line detection. To solve this problem of environmental adaptability of sea sky line detection in complex background, a sea sky line detection method is proposed based on the image edge phase encoding. The principle of bilateral filter is used to preserve the image edges and filter out the high frequency noise, and the edge response is enhanced by the Laplace of Gaussian. The edge phase encoding is enhanced and the noise of the system is restrained combined with the Gaussian-Kernel directional filter. Then, the optimized sea sky line is detected by the analysis of cumulative response intensity of scan line in phase component and the difference of pixel intensity distribution mode in sea sky region. The experimental results show that the proposed method can effectively detect the sea sky line in different complex backgrounds, and has low computation complexity and strong environmental adaptability.