A new generation of stationary orbit meteorological satellite Himawari-8 AHI is a multispectral earth observation sensor with high temporal and spatial resolutions, which is of great significance to the monitoring of sea fog. The vertical structure of the atmosphere at high-resolution can be detected by cloud-aerosol lidar with orthogonal polarization(CALIOP). CALIOP data were used to classify the observed objects into four types: sea-surface, sea fog, stratus and middle/high-cloud. These samples were then used to the spectral analysis of the Himawari-8 channels and the spectral properties were obtained. A daytime sea fog detection algorithm based on Himawari-8 data was proposed through the study of the spectral properties. Both CALIOP data and fog-monitoring product from NSMC were used to validate the detection results. Compared with the CALIOP fog detection, the probability of sea fog detection is 63.42%, and 71.54% of the missing samples are the misjudgements of sea surfaces. Based on the fog-monitoring product, the probability of sea fog detection is 89.7%. The results show that the proposed algorithm is feasible.