一种非显著场景图像的地理定位方法、装置及计算设备
By constructing a geospatial knowledge base and feature extraction model, the problem of accurate positioning of non-salient scene images was solved, achieving accurate geolocation of non-salient scene images and improving positioning accuracy and universality.
CN117493604BActive Publication Date: 2026-07-17BEIJING INST OF ENVIRONMENTAL FEATURES
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- BEIJING INST OF ENVIRONMENTAL FEATURES
- Filing Date
- 2023-11-17
- Publication Date
- 2026-07-17
AI Technical Summary
Technical Problem
Existing technologies struggle to accurately locate non-salient scene images that do not contain landmark references, obvious indicative background text, or strong regional human activity.
Method used
A geospatial knowledge base is constructed, including an index table, which stores the relationship between spatial semantic features, textual semantic features, and geographic locations. The geographic locations of non-salient scene images are determined through feature extraction models and image coding models.
Benefits of technology
It achieves accurate geolocation of images in non-salient scenes, improving the accuracy and universality of positioning in the absence of obvious geographical landmarks.
✦ Generated by Eureka AI based on patent content.
Smart Images
Figure CN117493604B_ABST
Abstract
本发明提供了一种非显著场景图像的地理定位方法、装置及计算设备，该方法包括：构建地理空间知识库；其中，地理空间知识库包括存储空间语义特征、文字语义特征和地理位置之间的关联关系的索引表；对待处理非显著场景图像进行特征提取，得到空间语义特征和文字语义特征；根据地理空间知识库、待处理非显著场景图像的空间语义特征和文字语义特征，确定待处理非显著场景图像的地理位置。本方案提供的非显著场景图像的地理定位方法提高了非显著场景图像的地理定位精度。
Need to check novelty before this filing date? Find Prior Art