Research On Complement Of Electronic Navigation Chart In Land Area Using Remote Sensing Image

China’s electronic nautical charts(ENC) production have basically achieved automation and the waters can be more accurately articulated. However, the Channel Data Acquisition Department’limited interest in the lands information leads to the loss of lands data and the serious lack of lands area in electronic chart. But, lands information play an important role in route optimization, management and landscaping. This paper based on the actual demand, using China’s GF-1 remote sensing image, focus on the river coastal land area feature extraction method and the fusion with original water data, expanding the land information in electronic chart.The main content of this paper is:(1) Source analysis and selectionThis paper compares Google Earth images and GF-1 images from the aspects of the data acquisition, data features, data quality and operability. At last, this paper choose GF-1 image as the source after careful consideration as GF-1 can meet the needs of this research and also can expand the application domain of domestic high-resolution satellite.(2) Research on typical feature extraction methodThe method is processed by data processing of principal component analysis(PCA), normalized differential vegetation index(NDVI), normalized differential water index (NDWI), and choose the appropriate feature element. This paper presents a unique four-feature extraction method of coastal land information which is suitable for electronic chart.(3) Fusion of remote sensing image feature information and original water dataThis paper used geographic information system operation platform, extracts the extraction feature information as vector form, unified the two data into the same coordinate system before loaded in ENC for display. Finally, complete the chart from editing and modifying and the data is consistent with the relevant specifications.This paper successfully extracts the building and vegetation information from the area of Huzhou (in Zhejiang Province), and the accuracy reached the actual application. GF-1 satellite images can be successfully fused with the original water data in electronic chart and obtained the ideal result. In summarize, this paper finds a new mapping method for land information in ENC, which is more convenient and faster than the traditional one, and can also be refreshed timely.