Refined glacial lake extraction in high Asia region by Deep Neural Network and Superpixel-based Conditional Random Field

Abstract. Remote sensing extraction of glacial lakes is an effective way of monitoring water body distribution and outburst disasters. At present, the lack of glacial lake datasets and the edge recognition problem of semantic segmentation networks lead to poor accuracy and inaccurate outlines of glacial lakes. Therefore, this study constructed a high-resolution dataset containing seven types of glacial lakes and proposed a refined glacial lake extraction method, which combines the LinkNet50 Network for rough extraction and Simple Linear Iterative Clustering (SLIC)-Dense Conditional Random Field (DenseCRF) for optimization. The results show that: 1) With Google Earth images of 0.52 m resolution in the study area, the Recall, Precision, F1 Score, and IoU of glacial lake extraction based on the proposed method are 96.52 %, 92.49 %, 94.46 %, and 90.69 %, respectively. 2) With the Google Earth images of 2.11 m resolution in the Qomolangma National Nature Reserve, 2300 glacial lakes with a total area of about 65.17 km² were detected by the proposed method. The area of the minimum glacial lake that can be extracted is about 160 m² (6×6 pixels). This method has advantages in small glacial lake extraction and refined outline detection, which can be applied to extracting glacial lakes in the high Asia region with high-resolution images.