A semantic segmentation method for high-resolution remote sensing images based on object enhancement

By adopting the improved resnet50 backbone network, ECA high-efficiency channel attention module, ASPP hollow pyramid pooling module and Transformer object feature learning module in high-resolution remote sensing image processing, the problem of difficulty in extracting spatial structure and object edge features and making full use of context information in the prior art is solved, and a higher segmentation accuracy is achieved.

CN116543152BActive Publication Date: 2025-04-18BEIJING JIUZHOU XINGTU TECHNOLOGY CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202310494644.X
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-05-05
Publication Date
2025-04-18
Estimated Expiration
2043-05-05

Smart Images

  • Figure CN116543152B_ABST
    Figure CN116543152B_ABST
Patent Text Reader

Abstract

The present invention relates to a high-resolution remote sensing image semantic segmentation method based on object enhancement, which belongs to the field of image processing. The method comprises the following contents: firstly, an efficient channel attention module is added to the backbone network to effectively capture the information of cross-channel interaction and improve the feature extraction capability of the backbone network. In the segmentation network, a spatial pyramid pooling module is used to further extract multi-scale information to solve the problem of large scale span between objects in remote sensing images. At the same time, an object feature learning module based on the transformer decoder structure is added, and the learned object features can enhance the pixel semantic information and improve the final segmentation accuracy.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Image Semantic Segmentation Method Based on Deep Full Convolutional Network and Conditional Random Field

    AU2020103901A4

  • Remote sensing image semantic segmentation method based on pyramid segmentation attention module

    CN113807210A