Journal of Marine Sciences ›› 2018, Vol. 36 ›› Issue (4): 43-52.DOI: 10.3969/j.issn.1001-909X.2018.04.006

Previous Articles     Next Articles

Research on coastline extraction method of WorldView-2 image combined with multi-spectral and texture features: Take the Cotton Island of Malaysia as an example

KAN Li-ping1, ZOU Ya-rong2,3, HU Zhuo-wei*1   

  1. 1. College of Resource Environment and Tourism, Capital Normal University, Beijing 100048, China;
    2. National Satellite Ocean Application Service, Ministry of Natural Resources, Beijing 100081, China;
    3. Key Laboratory of Space Ocean Remote Sensing and Application, SOA, Beijing 100081, China
  • Received:2018-08-31 Revised:2018-10-17 Online:2018-12-15 Published:2022-11-18

Abstract: Coastline is a fundamental element of the island and coastal zone research, which is of great significance to the exploitation and utilization of coastal resources. Remote sensing technology has become an important technical means for quickly extracting coastlines due to its macroscopic, multi-spectral and short cycle characteristics. In this study, the apparent reflectance of ground objects on both sides of the coastline was analyzed by using high-resolution remote sensing data WorldView-2, and the feature analysis was carried out to construct the coastline extraction index. On this basis, the shoreline of the Cotton Island, Malaysia was successfully extracted by further combining the texture feature extraction method. Then the accuracy of the extraction results was evaluated based on the lines matching method. The length error is -1.38%. When the buffer radius is 6 pixels, the integrity is 96.96% and the accuracy is 95.37%. The results show that this method could effectively extract the coastline and meet the demand for remote sensing survey.

Key words: coastline, apparent reflectance, extraction index, remote sensing

CLC Number: