Skip to content
naulon

Not checkedNot checked for notices yet.

A comparative study of impervious surface extraction using Sentinel-2 imagery

Junyi Chen, Suozhong Chen, Chao Yang and 3 more

European Journal of Remote Sensing · 2020

doi:10.1080/22797254.2020.1820383 (opens in a new tab)

We email you when a new notice or a question about this paper appears.

Abstract

The accuracy and efficiency of impervious surface extraction using different algorithms vary greatly, and algorithm applicability depends on the study area. Therefore, it is necessary to carry out a comparative study of different algorithms across different study areas. This study compared six impervious surface extraction indices (i.e., normalized difference built-up index (NDBI), index-based built-up index (IBI), biophysical composition index (BCI), combinational build-up index (CBI), combinational biophysical composition index (CBCI), and enhanced normalized difference impervious surfaces index (ENDISI)) using Sentinel-2 imagery in Fuxian Lake Basin, Shenzhen City, and Nanjing City. Three study areas with different geographical locations, climatic conditions and altitudes can test spatial heterogeneity of different indices. The results show that: (1) All indices could be used to extract impervious surface, but BCI and CBI were greatly disturbed by water bodies; (2) CBCI, IBI, and NDBI were influenced by study area, while ENDISI could be used across all three study areas; (3) ENDISI algorithm was the best among the six algorithms with a much higher separability degree and an overall accuracy of more than 91.00%. ENDISI can extract impervious surface quickly and accurately from Sentinel-2 imagery across different study areas, and can be well applied in the field of impervious surface change monitoring.

Abstract licence: CC BY-NC

References61

    40 more references, whose wording the paper’s licence keeps private.

    All 61 references

    Reads

    Read by people on this page

    None yet.

    Checked or read by AI through naulon

    None yet.

    A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.

    Are you an author?

    Is this your paper?

    Claim it to add your note and be paid when agents cite it. Free.