Fusion filtering multi-scale high-resolution remote sensing glacier extraction method

• Main Authors: ZHANG XIUZAI, ZHANG HAO, XIA CHENWEI
• Format: Patent
• Language: Chinese; English
• Published: 04.06.2024
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; PHYSICS

The invention provides a fusion filtering multi-scale high-resolution remote sensing glacier extraction method, which comprises the following steps of: firstly, performing data preprocessing on a high-resolution glacier remote sensing image to make a deep learning semantic segmentation data set, enriching the data set through a data enhancement means, and ensuring the robustness of model training; a gating multi-scale filter layer (G-MsFL) is designed to solve the problem that the recognition capability of scattered and small glaciers is insufficient. Different-scale extraction and feature fusion modes are provided for the model to capture fine glaciers, and useless feature information is effectively filtered through a gating mechanism; a parallel double-channel attention module (P-DAM) is designed for solving the problem that the glacier contour is blurred. Rich context information of a glacier boundary is coded as local features of a feature map, so that the feature expression ability is enhanced. And effec