Abstract:
Objectives In recent years, nighttime light images have been widely used in urban built-up area extraction. However, the accuracy of the extraction results is not satisfactory due to the low data resolution and the blooming effect.
Methods To solve this problem, the NTL(nighttime light) & POI(point of interest)composite index is proposed, which is based on the nighttime light images of Luojia 1-01 of China and NPP/VIIRS(national polar-orbiting partnership's visible infrared imaging radiometer suite) of the United States. The threshold method is used to extract the built-up areas before and after NTL&POI composite index processing.
Results Compared with the reference built-up areas, results extracted by NTL&POI composite index not only reflect the overall shape of the built-up areas but also retain details such as boundary information. The landscape pattern indexes indicate that the extracted results have better connectivity and complexity. Recall, precision and F1 score prove that NTL&POI composite index effectively improves the extraction accuracy of built-up areas, and achieves good results in main and non-main urban areas with different development degrees. Combined with POI data, the extraction accuracy of built-up areas of Luojia 1-01 increases by approximately 5% and that of NPP/VIIRS increases by about 3%.
Conclusions Compared with NPP/VIIRS, Luojia 1-01 ensures high precision and satisfying performance in terms of boundary and detail information of built-up areas, which brings it a broad application prospect.