Segmented Crop Mapping for Large Regions With Fewer Samples
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing crop mapping models face challenges in large regions due to differences in agricultural planting caused by topography, landforms, soil, and climate, poor adaptability, low accuracy in crop recognition, and high dependence on crop samples, leading to high costs and spatial limitations.
Innovation Solution
A method and system for crop mapping that involves geographically dividing regions by crop growth periods, establishing key growth period model libraries, constructing machine learning models using multiple algorithms, and performing product correction based on disaster information to reduce sample dependence and improve accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If existing crop mapping models are used across large regions, then crop mapping can be performed, but the models show poor adaptability due to great differences in agricultural planting caused by topography, landforms, soil, and climate
Solution Approach 1:
The patent divides the large study area into multiple sub-regions based on geographic, climatic, and agricultural characteristics. Each sub-region is trained separately with its own crop mapping model, allowing the system to adapt to local variations in topography, soil, and climate while maintaining overall reliability across the large region.
Solution Approach 2:
The patent implements location-specific model parameters and features for each sub-region, allowing the crop mapping system to capture local agricultural patterns and environmental characteristics. This enables high accuracy in each local area while the ensemble of regional models provides broad adaptability across the entire large region.
2Measurement precision
If high accuracy in crop recognition is achieved, then crop mapping quality improves, but the system becomes highly dependent on crop samples leading to high costs
Solution Approach 1:
The patent combines multiple data sources including satellite imagery, remote sensing data, climate information, and soil characteristics into a unified modeling framework. This integration allows the system to achieve high crop recognition accuracy by leveraging complementary information from multiple sources rather than relying solely on extensive ground sample data.
Solution Approach 2:
The patent introduces intermediate features and proxy variables that bridge the gap between available data and crop recognition targets. These intermediaries include derived spectral indices, climate-derived growth parameters, and soil moisture estimates, which enable accurate crop mapping with reduced direct dependence on ground truth samples.
3Area of stationary object
If crop mapping is performed across large regions with diverse agricultural planting, then comprehensive monitoring is achieved, but the complexity of the mapping system increases
Solution Approach 1:
The patent divides the complex large-region mapping problem into multiple manageable sub-problems by segmenting the study area into homogeneous sub-regions. Each sub-region can be processed independently with standardized procedures, reducing overall system complexity while maintaining comprehensive coverage across the large area through aggregation of regional results.
Data Source
AI summary
The present invention belongs to the technical field of crop mapping based on remote-sensing images, and relates to a method and system for crop mapping across large regions with low sample dependence. The method includes: acquiring remote sensing data, ground sample data, meteorological data, soil data, establishing geographically divided crop planting regions; establishing key growth period model libraries corresponding to individual crop regions; constructing machine learning models based on a plurality of machine learning algorithms, to obtain machine learning crop extraction models; selecting an optimal machine learning crop extraction model; acquiring a spatial crop distribution base map; performing product correction based on the disaster information; and acquiring a regional crop map using a target crop extraction model adapted for the disaster response. The present invention, achieve high-accuracy and large-scale crop mapping, and reduce the crop sample dependence of crop mapping.