Related Experiment Video
Updated: May 30, 2026
08:47
Computer Vision-Based Biomass Estimation for Invasive Plants
Published on: February 9, 2024
[Study on the method of recognizing abandoned farmlands based on multispectral remote sensing].
Wei-Fang Cheng1, Yi Zhou, Shi-Xin Wang
1The State Key Laboratory of Remote Sensing Science, Institute of Remote Sensing Application, Chinese Academy of Sciences, Beijing 100101, China. cwffang@163.com
Guang Pu Xue Yu Guang Pu Fen Xi = Guang Pu
|August 19, 2011
Summary
Abandoned farmland in China significantly impacts grain output. This study uses multispectral remote sensing data, specifically NDVI time-series analysis, to accurately identify abandoned farmlands, achieving 90% accuracy.
Area of Science:
- Agricultural Science
- Remote Sensing
- Geospatial Analysis
Context:
- Farmland abandonment is a significant issue in China, affecting grain production over the last two decades.
- Multispectral remote sensing offers efficient data acquisition for land use research.
Purpose:
- To develop and validate a method for extracting abandoned farmland in China using remote sensing data.
- To analyze the potential of NDVI time-series data for land use classification.
Summary:
- Utilized NDVI data from MODIS/Terra (2000-2009) and ALOS satellite imagery to characterize land use types.
- Analyzed NDVI time-series curves to identify patterns specific to abandoned farmland.
- Field surveys confirmed the accuracy of the remote sensing-based classification, reaching 90%.
Impact:
- Demonstrates the feasibility of using multispectral remote sensing for abandoned farmland identification.
- Provides a reliable method for monitoring land use changes and supporting agricultural policy in China.
