Flood detection method and device, equipment, storage medium and program product
By fusing SAR images with DEM data and combining them with multispectral images for flood detection, the problem of low flood detection accuracy in the prior art is solved, and higher precision flood detection is achieved.
CN119478658APending Publication Date: 2025-02-18AEROSPACE INFORMATION RES INST CAS
View PDF 0 Cites -1 Cited by
Patent Information
- Application Number
- CN202411382387.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-30
- Publication Date
- 2025-02-18
AI Technical Summary
Technical Problem
The accuracy of flood detection in the prior art is not high, mainly because the important role of background data in flood areas in flood detection is ignored.
Method used
By fusing the SAR image with the DEM data, the fused data is formed and inputted into the target model with the multi-spectral image, flood detection is performed using multimodal data. The target model includes a splicing module, an encoder, a decoder and a segmentation head. Through feature extraction, fusion and classification, it generates high-precision flood detection results.
Benefits of technology
By considering the important role of DEM data in flood detection, the accuracy of flood detection is improved, and the characteristics of floods in real scenes can be better captured.
✦ Generated by Eureka AI based on patent content.
Smart Images
Figure CN119478658A_ABST
Abstract
The invention provides a flood detection method, device and equipment, a storage medium and a program product, and belongs to the technical field of flood detection. The method comprises the following steps: inputting an SAR image, DEM data and a multispectral image of a target area into a target model to obtain a flood detection result of the target area output by the target model; the target model comprises a splicing module used for splicing the SAR image and the DEM data to obtain fused data; the first encoder is used for performing feature extraction on the fused data to obtain a first multi-level feature; the second encoder is used for performing feature extraction on the multispectral image to obtain a second multi-level feature; the decoder is used for fusing the first multi-level features and the second multi-level features to obtain multi-modal data; and the segmentation head is used for classifying the multi-modal data to obtain first data combined by the SAR image and the DEM data and second data based on the multi-spectral image. The flood detection precision can be improved.
Need to check novelty before this filing date? Find Prior Art