Situation awareness is the key foundation for intelligent ships to achieve autonomous navigation at sea. The light signal displayed by the ship at night is one of the most basic and important situational information. This paper proposed an effective ship lights recognition algorithm using image processing for maritime sailing ships. The algorithm first carries out the pre-processing of collected video images at night on the sea, and then makes the RGB color space of images transfer into the HSV. Set the color threshold based on the International Commission on Illumination (CIE) standard for the chromaticity of ship lights and perform color segmentation. Specific morphological open operations are employed to generate candidate regions, and finally realizes ship lights identification based on geometric property and regional pixel information. These results can provide a comprehensive guideline for intelligent vessels that use camera technology and computer vision algorithms for look-out instead of human in the future.

CICTP