Lightweight dual prediction branch semantic segmentation water body extraction deep learning method and system

By constructing a lightweight, dual-prediction-branch semantic segmentation deep learning model for water body extraction, Double-Net solves the problems of numerous model parameters and slow training in remote sensing images, achieving high-precision automated identification of water body information, especially improving accuracy in identifying small water bodies, and shortening the training time to 1 hour.

CN117911701BActive Publication Date: 2026-06-26HOHAI UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
HOHAI UNIV
Filing Date
2024-01-25
Publication Date
2026-06-26

AI Technical Summary

Technical Problem

Existing methods for extracting water bodies from remote sensing images suffer from numerous model parameters, slow training processes, difficulty in meeting real-time requirements, and insufficient accuracy in identifying small water bodies, especially in complex areas and small water bodies where there are missed detections.

Method used

We construct a lightweight deep learning model, Double-Net, for semantic segmentation and water extraction using dual prediction branches. It employs cascaded normalized convolutional blocks, residual-convolutional blocks, and dilated-convolutional blocks, combined with an adaptive learning rate training method. By fusing feature maps of different resolutions through dual prediction branches, we improve model accuracy and training efficiency.

Benefits of technology

It achieves high-precision automated identification of water body information, especially improving the accuracy of small water bodies identification. The model parameters are reduced to 1.5M, and the training time is shortened to about 1 hour, meeting the real-time requirements.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN117911701B_ABST
    Figure CN117911701B_ABST
Patent Text Reader

Abstract

The application discloses a lightweight dual-prediction branch semantic segmentation water body extraction deep learning method and system, first, the coding part of the lightweight dual-prediction branch semantic segmentation model is constructed, three convolution blocks in the coding structure are designed based on the visual perception, and high, medium and low resolution feature maps are extracted respectively; secondly, the decoding part of the lightweight dual-prediction branch semantic segmentation model is constructed, the dual-prediction branch is designed, the medium and low resolution feature maps are fused and used as one of the discriminant features, input into one of the prediction branches, the high resolution feature map is input into the other prediction branch, and the maximum confidence of the two prediction branches is used as the final prediction probability; finally, the model is trained by using a satellite remote sensing image labeled data set and the adaptive learning rate. The deep learning model provided in the application can effectively solve the water body extraction in a large range of satellite remote sensing images, and provide an intelligent method for dynamic change monitoring of water resources.
Need to check novelty before this filing date? Find Prior Art