一种遥感影像云量检测方法及系统
By using a cloud cover estimation model based on convolutional neural networks, cloud cover estimates are directly generated, solving the problem of high computational complexity in existing technologies. This enables fast and accurate cloud cover estimation in orbit, making it suitable for environments with limited on-board computing.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- WUHAN UNIV
- Filing Date
- 2022-12-19
- Publication Date
- 2026-07-17
AI Technical Summary
Existing on-orbit cloud detection methods suffer from high computational complexity and limited resources, making it difficult to achieve fast and accurate cloud cover estimation, especially in spaceborne environments where they are difficult to promote and use.
A cloud cover estimation model based on convolutional neural networks is adopted. The cloud cover estimate is directly generated through the feature extraction module and the cloud estimation module, skipping the cloud region segmentation step. The dataset is constructed using data from the GF1 remote sensing satellite, and the cloud mask generation module assists in supervising the model training, thereby reducing computational complexity.
It achieves fast and accurate cloud cover estimation in orbit, reduces model computational complexity, improves detection efficiency, and is suitable for environments with limited on-board computing.
Smart Images
Figure CN115908944B_ABST