Remote sensing image area correction method, system and device based on mixed pixel weighting and medium
CN116485878BActive Publication Date: 2026-07-03XIAN MICROELECTRONICS TECH INST
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- XIAN MICROELECTRONICS TECH INST
- Filing Date
- 2023-04-21
- Publication Date
- 2026-07-03
AI Technical Summary
Technical Problem
Existing technologies have significant errors in calculating crop planting area based on remote sensing images, especially in the edges of planting patches where misclassification is common, and they fail to effectively account for the influence of weather and instrument factors.
Method used
A remote sensing image area correction method based on mixed pixel weighting is adopted. By constructing a mixed pixel saturation dataset and an area error dataset, the weights are optimized using a genetic algorithm to build an area correction model and reduce errors.
Benefits of technology
This effectively reduces the error in calculating the area of remote sensing images, improves the accuracy of calculating the planting area of crops, and lowers the error rate.
✦ Generated by Eureka AI based on patent content.
Smart Images
Figure CN116485878B_ABST
Abstract
This invention discloses a method, system, device, and medium for remote sensing image area correction based on mixed pixel weighting. The method includes: classifying a remote sensing image dataset to obtain a classification result dataset S, and downloading a standard area dataset A; calculating the area based on a remote sensing image pixel area calculation model on the classification result dataset S to obtain an initial area dataset; calculating the initial area error based on the standard area data in the standard area dataset A and the initial area dataset to obtain an area error dataset; obtaining a mixed pixel saturation dataset based on the classification result dataset S; constructing a remote sensing image area correction model based on mixed pixel weighting based on the mixed pixel saturation dataset and the area error dataset; and obtaining the remote sensing image area based on the remote sensing image area correction model. This invention provides a more effective model, achieving area correction of remote sensing images and reducing the error rate of calculations based on classification results.
Need to check novelty before this filing date? Find Prior Art