Automatic labeling method for recommending information based on remote sensing image
By constructing an automated annotation model and utilizing remote sensing knowledge graphs for recommended annotation of remote sensing images, the problems of resource waste and long time in remote sensing image annotation are solved, and accurate annotation and automatic updates are achieved.
CN115934966BActive Publication Date: 2026-06-19BEIJING DATA INTELLIGENCE INFORMATION TECH CO LTD
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- BEIJING DATA INTELLIGENCE INFORMATION TECH CO LTD
- Filing Date
- 2023-01-03
- Publication Date
- 2026-06-19
Smart Images

Figure CN115934966B_ABST
Abstract
This invention provides an automatic annotation method based on remote sensing image recommendation information, belonging to the field of remote sensing image annotation. The method includes: constructing an automated annotation model, which comprises a recommendation module, an annotation module, and a feedback module, wherein the recommendation module includes a remote sensing knowledge graph; acquiring remote sensing samples and remote sensing tasks; inputting the remote sensing tasks into the recommendation module; using the remote sensing knowledge graph for recommendation to obtain recommended annotation items corresponding to the remote sensing tasks; then inputting the remote sensing samples into the annotation module; annotating the remote sensing samples based on the recommended annotation items to obtain annotation results; using the feedback module to evaluate the annotation results and obtain feedback information; and updating the annotation model based on the feedback information. This invention achieves accurate annotation, avoids resource waste, and shortens annotation time. The automated annotation model in this invention can be continuously updated to achieve automatic annotation results.
Need to check novelty before this filing date? Find Prior Art