Publication Date:
2020-04-03
Description:
The HY-1C satellite, as part of China’s optical satellite constellation for global oceanmonitoring, monitors the ocean and coastal environment by the three broad visible bands ofthe Coastal Zone Imager (CZI) instrument. However, as a result of the sensor instrument noise,the atmospheric environment during imaging, and the shooting angle, the satellite images often showuneven illumination and inconsistent color between neighboring images. In this paper, accordingto the characteristics of the HY-1C CZI instrument, we propose a color consistency processingframework for coastal zone images of Antarctica. First of all, the high-frequency and low-frequencyinformation of the image is separated by a statistical filter with simple clustering. The uneven lightingis then replaced by artificial lighting, which is globally uniform. Finally, the color difference betweenimages is corrected by a color transfer method. In order to evaluate the color consistency resultsquantitatively, a new quantitative evaluation method is proposed. The experimental results for thecoastal zone images of Antarctica show that the new processing framework can effectively eliminatethe unevenness in the lighting and color. The mosaic results show a good performance in consistentlighting and tones, and the lack of visible mosaic lines proves the effectiveness of the proposedmethod. The quantitative evaluation analysis confirms the superiority of the proposed method overthe Wallis method.
Electronic ISSN:
2072-4292
Topics:
Architecture, Civil Engineering, Surveying
,
Geography
Permalink