Land Use Classification via Satellite Temperature Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing land use databases are incomplete in geographic coverage, vary in data quality, and lack up-to-date information, making it difficult to accurately classify land use for weather and climate forecasting, especially in urban areas where building heights and materials impact local temperatures.
Innovation Solution
A method using high-resolution surface temperature data combined with machine learning models to classify land use development density levels, training models with labeled geographic data to analyze and classify areas lacking accurate land use data, thereby enhancing the accuracy of weather and climate predictions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If visible image data is used to indicate land use boundaries, then land use boundaries can be identified, but sufficient details for land use classification are lacking
Solution Approach 1:
The patent segments the land use classification problem into multiple components by using satellite temperature data to distinguish between different building material types (e.g., asphalt, concrete, metal) and intensity levels (low, medium, high) that cannot be differentiated by visible images alone. This segmentation allows for more precise classification of urban areas.
Solution Approach 2:
The patent introduces satellite temperature data as an intermediary measurement to bridge the information gap left by visible images. By measuring surface temperatures, the system can infer building material types and intensity levels, effectively mediating between limited visual data and comprehensive land use classification.
2Productivity
If existing land use databases are used, then some geographic coverage is provided, but the data is incomplete and not always up to date
Solution Approach 1:
The patent performs preliminary classification of urban areas using satellite temperature data before weather forecasting is conducted. By pre-classifying land use patterns and updating databases with current temperature-based classifications, the system ensures that the most recent land use information is available for improved forecast accuracy.
Solution Approach 2:
The patent establishes a feedback mechanism where satellite temperature measurements continuously update land use classifications. As new temperature data becomes available, the system refines its understanding of building material types and intensity levels, feeding this updated information back into the classification system to maintain current and accurate land use data.
3Measurement precision
If high-resolution temperature data is used to account for landscape variations, then weather model accuracy is improved, but data processing complexity increases
Solution Approach 1:
The patent applies local quality analysis by examining temperature variations at specific locations within urban areas to determine building material types and intensity levels. By focusing on local temperature characteristics rather than treating entire regions uniformly, the system achieves high-resolution land use classification that improves weather model accuracy for specific urban areas.
Data Source
AI summary
Classifying land use by receiving geographic data and land use data for a geographic area, receiving surface temperature data for the geographic area, mapping the geographic data and temperature data to a set of map grid cells, determining temperature statistics for each map grid cell, training a machine learning model according to the land use data and temperature statistics, and classifying land use for map grid cells of a different geographic area according to the machine learning model.