Siam-Gabor-ResNet Used for Crevasse Detection With Ground-Penetrating Radar Data

Crevasse detection is crucial for glacier and climate research, and provides essential guidance for activities in glacier regions. Ground-penetrating radar (GPR) and machine learning are used to automatically detect crevasse. In this study, a Siam-Gabor-ResNet deep learning framework is proposed to detect crevasse automatically using GPR data. A contrast learning with the Siamese network framework is proposed to improve the accuracy of crevasse detection, which aims to increase the feature similarity between crevasses while simultaneously enhancing the feature distinctiveness between crevasse and continuous snow layers. Additionally, a trainable Gabor-ResNet feature extraction module is built by integrating the Gabor filter bank into the ResNet network and used to further reduce the complexity of model training while extracting multiscale features in a real-time manner. Experiments are performed on the Greenland dataset and the 2015 McMurdo dataset, which illustrate the effectiveness of the proposed method. The average accuracy rate (AR) of crevasse detection reaches 94.38%, which can detect the narrowest crevasse (0.6 m) in two datasets, with an average detection time of only 6.8 ms. The experimental results show that the proposed method can detect crevasse in real time, automatically, and accurately.