ConvNeXt with Context-Weighted Deep Superpixels for High-Spatial-Resolution Aerial Image Semantic Segmentation
Open Access
Abstract
Semantic segmentation of high-spatial-resolution (HSR) aerial imagery is critical for applications such as urban planning and environmental monitoring, yet challenges, including scale variation, intra-class diversity, and inter-class confusion, persist. This study proposes a deep learning framework that integrates convolutional networks (CNNs) with context-enhanced superpixel generation, using ConvNeXt as the backbone for feature extraction. The framework incorporates two key modules, namely, a deep superpixel module (Spixel) and a global context modeling module (GC-module), which synergistically generate context-weighted superpixel embeddings to enhance scene–object relationship modeling, refining local details while maintaining global semantic consistency. The introduced approach achieves mIoU scores of 84.54%, 90.59%, and 64.46% on diverse HSR aerial imagery benchmark datasets (Vaihingen, Potsdam, and UV6K), respectively. Ablation experiments were conducted to further validate the contributions of the global context modeling module and deep superpixel modules, highlighting their synergy in improving segmentation results. This work facilitates precise spatial detail preservation and semantic consistency in HSR aerial imagery interpretation tasks, particularly for small objects and complex land cover classes.Keywords
Funding Information
- National Natural Science Foundation of China (42201300)
- Zhejiang Provincial Basic Public Welfare Research Project of China (LGN22C130016)
- Hangzhou Joint Fund of the Zhejiang Provincial Natural Science Foundation of China (LHZQN25D010002)
This publication has 63 references indexed in Scilit:
- Quantifying the impact of urban trees on land surface temperature in global citiesISPRS Journal of Photogrammetry and Remote Sensing, 2024
- Integrating physical model-based features and spatial contextual information to estimate building height in complex urban areasInternational Journal of Applied Earth Observation and Geoinformation, 2023
- Semi-supervised knowledge distillation framework for global-scale urban man-made object remote sensing mappingInternational Journal of Applied Earth Observation and Geoinformation, 2023
- Breaking the resolution barrier: A low-to-high network for large-scale high-resolution land-cover mapping using low-resolution labelsISPRS Journal of Photogrammetry and Remote Sensing, 2022
- Weakly supervised high spatial resolution land cover mapping based on self-training with weighted pseudo-labelsInternational Journal of Applied Earth Observation and Geoinformation, 2022
- UNetFormer: A UNet-like transformer for efficient semantic segmentation of remote sensing urban scene imageryISPRS Journal of Photogrammetry and Remote Sensing, 2022
- Fine identification of the supply–demand mismatches and matches of urban green space ecosystem services with a spatial filtering toolJournal of Cleaner Production, 2022
- A comprehensive review on recent applications of unmanned aerial vehicle remote sensing with various sensors for high-throughput plant phenotypingComputers and Electronics in Agriculture, 2021
- UAVid: A semantic segmentation dataset for UAV imageryISPRS Journal of Photogrammetry and Remote Sensing, 2020
- Survey on semantic segmentation using deep learning techniquesNeurocomputing, 2019