Urban buildings are extracted from high-resolution Earth observation (EO) images using semantic segmentation networks like U-Net and its successors. Each re-iteration aims to improve performance by employing a denser skip connection mechanism that harnesses multi-scale features for accurate object mapping. However, denser connections increase network parameters and do not necessarily contribute to precise segmentation. In this paper, we develop three dual skip connection mechanisms for three networks (U-Net, ResUnet, and U-Net3+) to selectively deepen the essential feature maps for improved performance. The three mechanisms are evaluated on feature maps of different scales, producing nine new network configurations. They are evaluated against their original vanilla configurations on four building footprint datasets of different spatial resolutions, including a multi-resolution (0.3+0.6+1.2m) dataset that we develop for complex urban environments. The evaluation revealed that densifying the large- and small-scale features in U-Net and U-Net3+ produce up to 0.905 F1, more than TransUnet (0.903) and Swin-Unet (0.882) in our new dataset with up to 19x fewer parameters. The results conclude that selectively densifying feature maps and skip connections enhances network performance without a substantial increase in parameters. The findings and the new dataset will contribute to the computer vision domain and urban planning decision processes.
ResUNet-a: a deep learning framework for semantic segmentation of remotely sensed data
Data61, CSIRO(澳大利亚联邦科学与工业研究组织数据61中心);CSIRO Agriculture & Food(澳大利亚联邦科学与工业研究组织农业与食品研究所);ICRAR, The University of Western Australia(西澳大利亚大学国际射电天文学研究中心)
高分辨率航空影像的场景理解对于各种遥感应用中的自动化监测任务具有重要意义。由于感兴趣的对象的像素值的类内方差大,类间方差小,这仍然是一项具有挑战性的任务。近年来,深度卷积神经网络已开始用于遥感应用,并展示了对象像素级分类的最新性能。\textcolor{black}{Here we propose a reliable framework for performant results for the task of semantic segmentation of monotemporal very high resolution aerial images. Our framework consists of a novel deep learning architecture, ResUNet-a, and a novel loss function based on the Dice loss. ResUNet-a uses a UNet encoder/decoder backbone, in combination with residual connections, atrous convolutions, pyramid scene parsing pooling and multi-tasking inference. ResUNet-a infers sequen…
Neighborhood Attention Makes the Encoder of ResUNet Stronger for Accurate Road Extraction
Simon Fraser University(西蒙弗雷泽大学);Jalpaiguri Government Engineering College(贾尔派古里政府工程学院);University of Waterloo(滑铁卢大学);Helmholtz-Zentrum Dresden-Rossendorf (HZDR)(亥姆霍兹德累斯顿罗森多夫研究中心(HZDR));Institute of Advanced Research in Artificial Intelligence (IARAI)(人工智能高级研究院(IARAI))
A knowledge-based data-driven (KBDD) framework for all-day identification of cloud types using satellite remote sensing
Southern University of Science and Technology(南方科技大学);Peng Cheng Laboratory(鹏城实验室);Eastern Institute of Technology(东方理工大学);Ningbo Institute of Digital Twin(宁波数字孪生研究院);Peking University(北京大学)
An Integrated Neighborhood and Scale Information Network for Open-Pit Mine Change Detection in High-Resolution Remote Sensing Images
China Aero Geophysical Survey & Remote Sensing Center for Nature Resources, China Geological Survey(中国地质调查局自然资源航空物探遥感中心);National Engineering Laboratory for Satellite Remote Sensing Applications, Aerospace Information Research Institute, Chinese Academy of Sciences(中国科学院空天信息创新研究院遥感卫星应用国家工程实验室)
DPSF-Net: A Dual-Prior Spatial-Frequency Network for Real-World Remote Sensing Image Dehazing
School of Software Engineering, Jinling Institute of Technology(金陵科技学院软件工程学院);School of Information Engineering, Yancheng Institute of Technology(盐城工学院信息工程学院)
IM2HEIGHT: Height Estimation from Single Monocular Imagery via Fully Residual Convolutional-Deconvolutional Network
Remote Sensing Technology Institute (IMF), German Aerospace Center (DLR)(德国航空航天中心遥感技术研究所(IMF));Signal Processing in Earth Observation (SiPEO), Technical University of Munich (TUM)(慕尼黑工业大学地球观测信号处理研究所(SiPEO))
Cloud Removal With PolSAR-Optical Data Fusion Using A Two-Flow Residual Network
Aerospace Information Research Institute, Chinese Academy of Sciences(中国科学院空天信息创新研究院);International Research Center of Big Data for Sustainable Development Goals(国际可持续发展大数据研究中心);College of Resources and Environment, University of Chinese Academy of Sciences(中国科学院大学资源与环境学院)
A Cross-Hierarchical Difference Feature Fusion Network Based on Multiscale Encoder-Decoder for Hyperspectral Change Detection
School of Information and Communication Engineering, Hainan University(海南大学信息与通信工程学院);Department of Artificial Intelligence, School of Information and Communication Engineering, Hainan University(海南大学信息与通信工程学院人工智能系);Department of Management Information Systems, King Faisal University(国王法赫德大学管理信息系统系)