Field Patch Extraction Based on High-Resolution Imaging and U2-Net++ Convolutional Neural Networks
Open Access
- 10 October 2023
- journal article
- research article
- Published by MDPI AG in Remote Sensing
- Vol. 15 (20) , 4900
Abstract
Accurate extraction of farmland boundaries is crucial for improving the efficiency of farmland surveys, achieving precise agricultural management, enhancing farmers’ production conditions, protecting the ecological environment, and promoting local economic development. Remote sensing and deep learning are feasible methods for creating large-scale farmland boundary maps. However, existing neural network models have limitations that restrict the accuracy and reliability of agricultural parcel extraction using remote sensing technology. In this study, we used high-resolution satellite images (2 m, 1 m, and 0.8 m) and the U2-Net++ model based on the RSU module, deep separable convolution, and the channel-spatial attention mechanism module to extract different types of fields. Our model exhibited significant improvements in farmland parcel extraction compared with the other models. It achieved an F1-score of 97.13%, which is a 7.36% to 17.63% improvement over older models such as U-Net and FCN and a more than 2% improvement over advanced models such as DeepLabv3+ and U2-Net. These results indicate that U2-Net++ holds the potential for widespread application in the production of large-scale farmland boundary maps.Keywords
Funding Information
- Water Conservancy Science and Technology Project of Jiangsu Province (2021081, JZ110145B0012021)
- Hunan Province Water Conservancy Science and Technology Project (2021081, JZ110145B0012021)
- Three Gorges Follow-up Work (2021081, JZ110145B0012021)
- Basic scientific research business fund of the Chinese Academy of Water Sciences (2021081, JZ110145B0012021)
- Key Technology Research (2021081, JZ110145B0012021)
This publication has 30 references indexed in Scilit:
- Fully Convolutional Networks for Semantic SegmentationIEEE Transactions on Pattern Analysis and Machine Intelligence, 2016
- Holistically-Nested Edge Detection2015
- An Automated Method for Annual Cropland Mapping along the Season for Various Globally-Distributed Agrosystems Using High Spatial and Temporal Resolution Time SeriesRemote Sensing, 2015
- Image Segmentation Based on Constrained Spectral Variance Difference and Edge PenaltyRemote Sensing, 2015
- Automated crop field extraction from multi-temporal Web Enabled Landsat DataRemote Sensing of Environment, 2014
- Field-based sub-boundary extraction from remote sensing imagery using perceptual groupingISPRS Journal of Photogrammetry and Remote Sensing, 2013
- The yield gap of global grain production: A spatial analysisAgricultural Systems, 2010
- Object based image analysis for remote sensingISPRS Journal of Photogrammetry and Remote Sensing, 2010
- Minimization of Region-Scalable Fitting Energy for Image SegmentationIEEE Transactions on Image Processing, 2008
- Segmenting multispectral Landsat TM images into field unitsIEEE Transactions on Geoscience and Remote Sensing, 2002