MCD-Net: A Lightweight Deep Learning Baseline for Optical-Only Moraine Segmentation

2026-01-05

arXiv.org

AI Summary

This study addresses the limitations of existing automated moraine segmentation methods, which suffer from low contrast in optical imagery and the scarcity of high-resolution digital elevation models (DEMs). To overcome these challenges, the authors present the first large-scale, purely optical moraine segmentation dataset and propose MCD-Net—a lightweight architecture integrating a MobileNetV2 encoder, CBAM attention mechanism, and DeepLabV3+ decoder. Operating solely on high-resolution optical remote sensing imagery, the model achieves efficient segmentation with a mean Intersection over Union (mIoU) of 62.3% and a Dice coefficient of 72.8% on the test set. Moreover, it reduces computational costs by over 60% compared to deeper networks, demonstrating the feasibility of using only optical data for moraine delineation and offering a deployable baseline solution for resource-constrained environments.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Mixture of Experts (MoE)
• Data Mining & Knowledge Management: Mining of Visual, Multimedia & Multimodal Data

Application Category

• Web Mining and Content Analysis: Large pretrained models with web data
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

Glacial segmentation is essential for reconstructing past glacier dynamics and evaluating climate-driven landscape change. However, weak optical contrast and the limited availability of high-resolution DEMs hinder automated mapping. This study introduces the first large-scale optical-only moraine segmentation dataset, comprising 3,340 manually annotated high-resolution images from Google Earth covering glaciated regions of Sichuan and Yunnan, China. We develop MCD-Net, a lightweight baseline that integrates a MobileNetV2 encoder, a Convolutional Block Attention Module (CBAM), and a DeepLabV3+ decoder. Benchmarking against deeper backbones (ResNet152, Xception) shows that MCD-Net achieves 62.3% mean Intersection over Union (mIoU) and 72.8% Dice coefficient while reducing computational cost by more than 60%. Although ridge delineation remains constrained by sub-pixel width and spectral ambiguity, the results demonstrate that optical imagery alone can provide reliable moraine-body segmentation. The dataset and code are publicly available at https://github.com/Lyra-alpha/MCD-Net, establishing a reproducible benchmark for moraine-specific segmentation and offering a deployable baseline for high-altitude glacial monitoring.

Problem

• moraine segmentation
• optical imagery
• glacial monitoring
• automated mapping
• weak optical contrast

Innovation

• MCD-Net
• moraine segmentation
• lightweight deep learning
• optical-only imagery
• Convolutional Block Attention Module