Alphaearth foundations and machine learning based future year land use and land cover raster data downscaling method and system

By using AlphaEarth Foundations' Google Satellite Embedding dataset and machine learning models, the probability prediction of land use conversion is automated, which solves the problem of relying on manual rules and auxiliary data in existing technologies. It realizes a stable and reproducible downscaling method for land use and cover raster data in future years, and improves the physical meaning and spatial distribution rationality of the results.

CN121388828BActive Publication Date: 2026-06-23BEIJING NORMAL UNIVERSITY
2 Cites 0 Cited by

Patent Information

Application Number
CN202511427113.4
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2025-09-30
Publication Date
2026-06-23
Estimated Expiration
2045-09-30

AI Technical Summary

Technical Problem

Existing technologies for downscaling land use and cover raster data in future years rely on manually defined rules and difficult-to-obtain auxiliary data, resulting in unstable output quality. Furthermore, land use conversion rules lack physical meaning and are prone to jaggedness and bias.

Method used

Using the Google Satellite Embedding dataset and machine learning models from AlphaEarth Foundations, pixel-level embedding vectors are extracted through deep neural networks. Combined with random forest models and Bayesian optimization, an optimization model is constructed to automatically predict land use conversion probabilities, generate optimal land use allocation results, and reduce reliance on high-cost data and expert experience.

Benefits of technology

It achieves automated optimization of land use and cover raster data for future years, with stable and reproducible results, reduces jaggedness, improves the scientific and practical value of simulation results, and meets the overall balance of macro-level land demand.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The application provides a land use and cover grid data downscaling method and system based on AlphaEarth Foundations and machine learning in future years, which identifies coarse scale pixels in which land classes change in future years by comparing coarse scale data in future years and a base year; maps the coarse scale pixels to fine scale data in the base year to determine a target area; counts the number of fine scale pixels of each land class in the target area to obtain the net increase of fine scale pixels of each land class in future years; in the target area, generates a conversion probability corresponding to each land class for each fine scale pixel by using a land class conversion probability prediction model; constructs an optimization model with the sum of the conversion probabilities corresponding to the land classes to which all fine scale pixels in the target area are assigned as an objective function, and solves the optimization model to obtain an optimal land class assignment result of the fine scale pixels in the target area; and fuses the optimal land class assignment result with the fine scale data in the base year to generate fine scale land use and cover grid data in future years after downscaling.
Need to check novelty before this filing date? Find Prior Art