MMDLG: A lightweight multi-scale spatial MoE framework for image-based riparian vegetation measurement and instance segmentation on unmanned edge devices
Authors:
- Chen LiCLChen Li
- Jiannan ZhangJZJiannan Zhang
- Xinyu WangXWXinyu Wang
- Kaiqi WangKWKaiqi Wang
- Liu JianhuaLJLiu Jianhua
- 4 November 2026
- journal article
- Published by Elsevier in Measurement
- Vol. 293
Abstract
No abstract availableFunding Information
- China Institute of Water Resources and Hydropower Research
This publication has 36 references indexed in Scilit:
- Hydrodynamics and Sediment Transport Capacity of Vegetated Rivers Covered by Multiple Plants Species: A Reach-Scale AnalysisWater Resources Management, 2026
- Urban street tree species identification and factor interpretation model based on natural imagesUrban Forestry & Urban Greening, 2024
- YOLOv9: Learning What You Want to Learn Using Programmable Gradient Information2024
- ConvNeXt based semi-supervised approach with consistency regularization for weeds classificationExpert Systems with Applications, 2023
- An automatic approach for tree species detection and profile estimation of urban street trees using deep learning and Google street view imagesISPRS Journal of Photogrammetry and Remote Sensing, 2022
- Conditional Convolutions for Instance Segmentation2020
- Using remote sensing to characterize riparian vegetation: A review of available tools and perspectives for managersJournal of Environmental Management, 2020
- How vegetation can aid in coping with river management challenges: A brief reviewEcohydrology & Hydrobiology, 2018
- Factors driving the relationships between vegetation and soil properties in the Yellow River Delta, ChinaCATENA, 2018
- A review on the sustainability of constructed wetlands for wastewater treatment: Design and operationBioresource Technology, 2015