Urban land utilization mapping method based on multi-modal data collaborative awareness

By using a multimodal data collaborative perception method, an urban land use mapping framework was constructed, which solved the problem of sparse or unevenly distributed multi-source data, achieved high-precision and robust urban land use identification, and improved the discriminative power and identification accuracy of multi-source geographic features at the plot level.

CN122115756APending Publication Date: 2026-05-29CHINA UNIV OF GEOSCIENCES (WUHAN)
0 Cites 0 Cited by

Patent Information

Application Number
CN202610055118.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-01-15
Publication Date
2026-05-29

AI Technical Summary

Technical Problem

Existing technologies struggle to achieve high-precision and robust urban land use identification under conditions of uneven or sparse distribution of multi-source data. Furthermore, the fusion methods for multi-source data are difficult to effectively handle significant differences in spatial scale, distribution patterns, and representation space, which can easily introduce information redundancy and semantic confusion.

Method used

By constructing a multimodal data collaborative perception method, including land parcel division and spectral feature extraction based on road networks, setting street view sampling points to obtain street view perception features, constructing a heterogeneous graph structure by combining semantic features of points of interest, using graph attention networks to process spatial topological relationships and semantic associations, and performing semantic alignment and fusion through a fully sparse topic model to generate a fused land parcel feature representation.

Benefits of technology

Under conditions of sparse or unevenly distributed multi-source data, more accurate, consistent, and well-generalized refined identification of urban land use was achieved, improving the discriminative power of multi-source geographic features at the plot level and reducing semantic confusion and redundant information.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention belongs to the technical field of deep learning and remote sensing image processing, and particularly discloses a multi-modal data collaborative perception-based urban land utilization mapping method, which comprises the following steps of: performing plot division on a high-resolution remote sensing image of a research area to obtain an irregular urban plot and remote sensing spectral characteristics thereof; streetscape sampling points are set along the road network, streetscape perception features are extracted, and interest point semantic features are extracted; constructing a heterogeneous graph structure based on street scene sampling points, street scene perception features and interest point semantic features falling into the same plot, and then extracting plot-level multi-source geographic features; inputting the remote sensing spectral features and the plot-level multi-source geographic features into a full sparse topic model for semantic alignment and fusion, and generating fused plot feature representation; and performing classification based on the fused land parcel feature representation, and outputting an urban land utilization mapping result. According to the method, high-precision and high-robustness urban land utilization identification can be realized under the condition that the multi-source data distribution is non-uniform or sparse.
Need to check novelty before this filing date? Find Prior Art