Remote-sensing image rotating target detection method
A remote sensing image and target detection technology, applied in the field of aerial image target detection, can solve the problems of insufficient robustness, insufficient use of scene and target semantic information, inability to predict the coordinates of rotated quadrilaterals, etc., to reduce interference and improve accuracy.
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment
[0032] Example: A remote sensing image rotating target detection method, which proposes a new target detection network (GLS-Net), based on the Fast R-CNN network, combining the saliency algorithm RC with the feature pyramid network and adopts The saliency pyramid is constructed based on the saliency algorithm and neural network, and on the basis of the saliency pyramid, the global attention network branch is proposed, and the global semantic constraint network based on the channel attention mechanism is used to extract the depth features of the scene, and then on this basis, A fast feature fusion strategy based on the local target information of the saliency pyramid and the global semantic information optimized by the channel attention mechanism is proposed. Finally, an angle-sensitive IoU algorithm is proposed and used to optimize the non-maximum suppression algorithm during training. Effect to obtain a more accurate five-parameter rotating frame representation;
[0033] Specifi...
PUM
Abstract
Description
Claims
Application Information
- R&D Engineer
- R&D Manager
- IP Professional
- Industry Leading Data Capabilities
- Powerful AI technology
- Patent DNA Extraction
Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.
© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com