| 170 | 0 | 121 |
| 下载次数 | 被引频次 | 阅读次数 |
针对特定平台下遥感影像分割、分类应用，提出了一种基于ARM架构与Docker容器化部署的Swin-Transformer遥感影像云检测方法。通过构建无符号16位的图像-标签样本，保持地物的光谱细节不被压缩丢失，与传统的8位自然图像相比，提升了云与雪高亮类别的可分性和检测精度。同时，针对ARM架构硬件及操作系统，采用基于Docker容器化技术的跨平台部署方案，实现算法环境的一致性封装与灵活迁移。数据实验表明，利用基于ImageNet-1k样本预训练的Swin-Transformer模型进行小块推理并添加精细化调整进行模型迭代，结合模型迭代的主动学习策略，提升了复杂场景下的地物分类准确率，同时基于ARM的Docker部署方案保持了跨平台的兼容性，为特定环境中的遥感智能解译提供了可行技术路径。
Abstract：For remote sensing image segmentation and classification applications on specific platforms, a Swin-Transformer-based remote sensing image cloud detection method is proposed, which is deployed using ARM architecture and Docker containerization.By constructing unsigned 16-bit image-label samples, the spectral details of objects are preserved without compression loss.Compared to traditional 8-bit natural images, this approach improves the separability and detection accuracy of high-brightness categories such as clouds and snow.Additionally, for ARM architecture hardware and operating systems, a cross-platform deployment solution based on Docker containerization technology is adopted to achieve consistent encapsulation and flexible migration of the algorithm environment.Data experiments show that using a Swin-Transformer model pre-trained on ImageNet-1k samples for small-block inference, combined with fine-tuning for model iteration and an active learning strategy for model iteration, the object classification accuracy in complex scenarios is improved.Meanwhile, the ARM-based Docker deployment solution maintains cross-platform compatibility, providing a feasible technical path for remote sensing intelligent interpretation in specific environments.
[1] ZHANG Y C,ROSSOW W B,LACIS A A,et al.Calculation of Radiative Fluxes from the Surface to Top of Atmosphere Based on ISCCP and Other Global Data Sets:Refinements of the Radiative Transfer Model and the Input Data[J].Journal of Geophysical Research,2004,109(D19):2003JD04457.
[2] 窦世卿，宋莹莹，徐勇，等.基于随机森林的高分影像分类及土地利用变化检测[J].无线电工程，2021,51(9):901-908.
[3] 冀汶莉，王佳豪，王新伟.基于 LoRa 的农业大田土壤多参数监测系统设计[J].无线电工程，2023,53 (2):456-464.
[4] 王斯正，任雍，陈俊，等.基于空分天线的雷达大气目标探测技术研究[J].无线电工程，2024,54(11):2624-2632.
[5] 张福林，王思逸，彭望，等.基于改进Mask R-CNN的水电站水下建筑物缺陷检测[J].无线电工程，2025,55(5):966-974.
[6] QIU S,ZHU Z,HE B B.Fmask 4.0:Improved Cloud and Cloud Shadow Detection in Landsats 4-8 and Sentinel-2 Imagery[J].Remote Sensing of Environment,2019,231:111205.
[7] MAIN-KNORN M,PFLUG B,LOUIS J,et al.Sen2Cor for Sentinel-2[EB/OL].[2025-05-10].https://www.researchgate.net/publication/320231869_Sen2Cor_for_Sentinel-2.
[8] WANG J,YANG D D,CHEN S L,et al.Automatic Cloud and Cloud Shadow Detection in Tropical Areas for Planetscope Satellite Images[J].Remote Sensing of Environment,2021,264:112604.
[9] XU K,GUAN K,PENG J,et al.Deepmask:An Algorithm for Cloud And Cloud Shadow Detection In Optical Satellite Remote Sensing Images Using Deep Residual Network[EB/OL].(2019-11-09)[2025-05-11].https://arxiv.org/abs/1911.03607.
[10] MOHAJERANI S,SAEEDI P.Cloud-Net:An End-to-End Cloud Detection Algorithm for Landsat 8 Imagery[C]∥IGARSS 2019-2019 IEEE International Geoscience and Remote Sensing Symposium.Yokohama:IEEE,2019:1029-1032.
[11] YIN Z,LING F,FOODY G M,et al.Cloud Detection In Landsat-8 Imagery in Google Earth Engine Based on a Deep Convolutional Neural Network[EB/OL].[2025-05-11].https://arxiv.org/pdf/2006.10358v1.
[12] 隋淞蔓，夹尚丰，胡学谦.统一样本云检测技术在GF-6 WFV上的改进与应用[J].遥感学报，2022,26(4):646-656.
[13] GUO J H,YANG J Y,YUE H J,et al.Unsupervised Domain Adaptation for Cloud Detection Based on Grouped Features Alignment and Entropy Minimization[J].IEEE Transactions on Geoscience and Remote Sensing,2021,60:1-13.
[14] CHEN K,XIA M,LIN H F,et al.Multiscale Attention Feature Aggregation Network for Cloud and Cloud Shadow Segmentation [J].IEEE Transactions on Geoscience and Remote Sensing,2023,61:1-16.
[15] LU C,XIA M,QIAN M,et al.Dual-branch Network for Cloud and Cloud Shadow Segmentation[J].IEEE Transactions on Geoscience and Remote Sensing,2022,60:1-12.
[16] DONG J W,WANG Y H,YANG Y,et al.MCDNet:Multilevel Cloud Detection Network for Remote Sensing Images Based on Dual-perspective Change-guided and Multi-scale Feature Fusion[J].International Journal of Applied Earth Observation and Geoinformation,2024,129:103820.
[17] FOGA S,SCARAMUZZA P L,GUO S,et al.Cloud Detection Algorithm Comparison and Validation for Operational Landsat Data Products[J].Remote Sensing of Environment,2017,194:379-390.
[18] LI Z W,SHEN H F,LI H F,et al.Multi-feature Combined Cloud and Cloud Shadow Detection in GaoFen-1 Wide Field of View Imagery[J].Remote Sensing of Environment,2017,191:342-358.
[19] FAN X,CHANG H,HUO L Z,et al.GF-1/6 Satellite Pixel-by-Pixel Quality Tagging Algorithm[J].Remote Sensing 2023,15:1955.
[20] KMIEC S,WONG J,JACOBSEN H A,et al.A Comparison of ARM Against x86 for Distributed Machine Learning Workloads[C]//Performance Evaluation and Benchmarking for the Analytics Era.Munich:Springer,2017:164-184.
基本信息:
中图分类号:TP751
引用信息:
[1]陆俊南,戴山,胡昌苗.基于ARM架构与Docker的Swin-Transformer遥感影像云检测方法研究[J].无线电工程,2025,55(12):2373-2384.
基金信息:
国家重点研发计划(2024YFD1500802)~~
2025-08-25
2025-08-25
2025-08-25