An attention-fused network for semantic segmentation of very-high-resolution remote sensing imagery
نویسندگان
چکیده
Semantic segmentation is an essential part of deep learning. In recent years, with the development remote sensing big data, semantic has been increasingly used in sensing. Deep convolutional neural networks (DCNNs) face challenge feature fusion: very-high-resolution image multisource data fusion can increase network's learnable information, which conducive to correctly classifying target objects by DCNNs; simultaneously, high-level abstract features and low-level spatial improve classification accuracy at border between objects. this paper, we propose a multipath encoder structure extract inputs, attention-fused block module fuse features, refinement features. Furthermore, novel network architecture, named (AFNet). Based on our AFNet, achieve state-of-the-art performance overall 91.7% mean F1 score 90.96% ISPRS Vaihingen 2D dataset 92.1% 93.44% Potsdam dataset.
منابع مشابه
An Object-Based Semantic Classification Method for High Resolution Remote Sensing Imagery Using Ontology
Haiyan Gu 1,*, Haitao Li 1, Li Yan 2, Zhengjun Liu 1, Thomas Blaschke 3 and Uwe Soergel 4 1 Institute of Photogrammetry and Remote Sensing, Chinese Academy of Surveying and Mapping, 28 Lianhuachi Road, Beijing 100830, China; [email protected] (H.L.); [email protected] (Z.L.) 2 School of Geodesy and Geomatics, Wuhan University, Luojiashan, Wuhan 430072, China; [email protected] 3 Department of G...
متن کاملA new process for the segmentation of high resolution remote sensing imagery
International Journal of Remote Sensing Publication details, including instructions for authors and subscription information: http://www.informaworld.com/smpp/title~content=t713722504 A new process for the segmentation of high resolution remote sensing imagery Z. Chen ab; Z. Zhao a; P. Gong b; B. Zeng c a Institute of Remote Sensing Applications, Chinese Academy of Science. Beijing 100101. Peop...
متن کاملSegmentation Improvement of High Resolution Remote Sensing Images based on superpixels using Edge-based SLIC algorithm (E-SLIC)
The segmentation of high resolution remote sensing images is one of the most important analyses that play a significant role in the maximal and exact extraction of information. There are different types of segmentation methods among which using superpixels is one of the most important ones. Several methods have been proposed for extracting superpixels. Among the most successful ones, we can r...
متن کاملSegmentation for High-Resolution Optical Remote Sensing Imagery Using Improved Quadtree and Region Adjacency Graph Technique
An approach based on the improved quadtree structure and region adjacency graph for the segmentation of a high-resolution remote sensing image is proposed in this paper. In order to obtain the initial segmentation results of the image, the image is first iteratively split into quarter sections and the quadtree structure is constructed. In this process, an improved fast calculation method for st...
متن کاملHigh Resolution Sensing and Anisotropic Segmentation for SAR Imagery
The purpose of this note is to introduce new methodologies for high resolution image processing and knowledge-based segmentation for SAR imagery. These techniques could also have a major impact on problems in radar and remote sensing where modern mathematical tools and optimization techniques are likely to advance significantly the current state of the art.
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Isprs Journal of Photogrammetry and Remote Sensing
سال: 2021
ISSN: ['0924-2716', '1872-8235']
DOI: https://doi.org/10.1016/j.isprsjprs.2021.05.004