Automatic Extraction of the Basal Channel Based on Neural Network

The basal channel is a morphological feature that widely exists at the bottom of ice shelves which mainly formed by incision warm water base shelves. study can contribute to further understanding impact global warming on polar In this manuscript, depression caused surface an shelf used confirm and extract channel's position. automatic acquisition realized machine learning. To more efficiently minimize subjectivity visual interpretation, neural network was applied for first time, effectively improved accuracy extraction. We took manual extraction results channels as training set input from model. U-NET model classify 79NG area, 73.58%. Then, trained extended Peterman Ryder located in Greenland. accuracies were 75.54% 72.70%, respectively.