面向遥感图像场景分类的GLFFNet模型

doi:10.11947/j.AGCS.2023.20220286

测绘学报 ›› 2023, Vol. 52 ›› Issue (10): 1693-1702.doi: 10.11947/j.AGCS.2023.20220286

面向遥感图像场景分类的GLFFNet模型

王威¹, 邓纪伟¹, 王新¹, 李智勇², 袁平³

1. 长沙理工大学计算机与通信工程学院, 湖南长沙 410114;
2. 湖南神帆科技有限公司, 湖南长沙 410011;
3. 长沙市竟网信息科技有限公司, 湖南长沙 410010

收稿日期:2022-05-05 修回日期:2023-02-22 发布日期:2023-10-31
通讯作者: 王新 E-mail:wangxin@csust.edu.cn
作者简介:王威(1974-),男,博士,教授,博士生导师,研究方向为计算机视觉和模式识别。E-mail:wangwei@csust.edu.cn
基金资助:
湖南省重点研究开发项目(2020SK2134);湖南省自然科学基金项目(2019JJ80105;2022JJ30625);长沙市科技计划项目(kq2004071)

GLFFNet model for remote sensing image scene classification

WANG Wei¹, DENG Jiwei¹, WANG Xin¹, LI Zhiyong², YUAN Ping³

1. School of Computer and Communication Engineering, Changsha University of Science and Technology, Changsha 410114, China;
2. Hunan Shenfan Science and Technology Limited Company, Changsha 410011, China;
3. Changsha Jingwang Information and Technology Limited Company, Changsha 410010, China

Received:2022-05-05 Revised:2023-02-22 Published:2023-10-31
Supported by:
Key Research and Development Project of Hunan Province (No. 2020SK2134);Hunan Natural Science Foundation Project (Nos. 2019JJ80105;2022JJ30625);Changsha Science and Technology Planning Project (No. kq2004071)

摘要/Abstract

摘要： 传统场景分类模型不能轻量高效地完成遥感图像中多尺度的关键特征提取,深度学习方法普遍存在计算量大、收敛速度慢等缺点。针对以上问题,本文充分利用CNN结构和Transformer结构对不同尺度特征的提取能力,提出了一种全局-局部特征提取模块(global and local features fused block,GLFF),并基于此模块设计了一个轻量级遥感图像场景分类模型(GLFFNet),该模型具有较好的局部信息和全局信息提取能力。为了验证GLFFNet的有效性,本文使用开源遥感图像数据集RSSCN7与SIRI-WHU测试GLFFNet与其他深度学习网络的复杂度和识别能力。最终,GLFFNet在RSSCN7与SIRI-WHU数据集上分别取得了高达94.82%和95.83%的识别准确率,优于其他先进的模型。

关键词: 遥感图像, 场景分类, 卷积神经网络, Transformer, GLFFNet模型

Abstract: Traditional scene classification models cannot perform multi-scale key feature extraction in remote sensing images in a lightweight and efficient manner. Deep learning methods generally have shortcomings such as large amount of calculation and slow convergence speed. In view of the above problems, this paper makes full use of the ability of CNN structure and Transformer structure to extract features at different scales, and proposes a feature extract module, named global and local features fused (GLFF) block. Based on this module, a lightweight remote sensing image scene classification model, GLFFNet, is designed, which has better local information and global information extraction ability. In order to verify the effectiveness of GLFFNet, this paper uses the open-source remote sensing image datasets RSSCN7 and SIRI-WHU to verify the complexity and recognition ability of GLFFNet and other deep learning networks. Finally, GLFFNet achieves recognition accuracy of up to 94.82% and 95.83% on RSSCN7 and SIRI-WHU datasets, respectively, which is better than other state-of-the-art models.

Key words: remote sensing image, scene classification, convolutional neural network, Transformer, GLFFNet model

中图分类号:

P237

王威, 邓纪伟, 王新, 李智勇, 袁平. 面向遥感图像场景分类的GLFFNet模型[J]. 测绘学报, 2023, 52(10): 1693-1702.

WANG Wei, DENG Jiwei, WANG Xin, LI Zhiyong, YUAN Ping. GLFFNet model for remote sensing image scene classification[J]. Acta Geodaetica et Cartographica Sinica, 2023, 52(10): 1693-1702.

参考文献

[1] 龚健雅,张觅,胡翔云,等.智能遥感深度学习框架与模型设计[J].测绘学报, 2022, 51(4):475-487. DOI:10.11947/j.AGCS.2022. 20220027. GONG Jianya, ZHANG Mi, HU Xiangyun, et al. The design of deep learning framework and model for intelligent remote sensing[J]. Acta Geodaetica et Cartographica Sinica, 2022, 51(4):475-487. DOI:10.11947/j.AGCS.2022. 20220027.
[2] 乔星星,施文灶,刘芫汐,等.基于ResNet双注意力机制的遥感图像场景分类[J].计算机系统应用, 2021, 30(8):243-248. QIAO Xingxing, SHI Wenzao, LIU Yuanxi, et al. Remote sensing image scene classification based on ResNet and dual attention mechanism[J]. Computer Systems&Applications, 2021, 30(8):243-248.
[3] WANG Wang, LI Shaochun, WANG Wang, et al. A simple deep learning network for classification of 3 D mobile LiDAR point clouds[J]. Journal of Geodesy and Geoinformation Science, 2021, 4(3):49-59.
[4] 边小勇,费雄君,穆楠.基于尺度注意力网络的遥感图像场景分类[J].计算机应用, 2020, 40(3):872-877. BIAN Xiaoyong, FEI Xiongjun, MU Nan. Remote sensing image scene classification based on scale-attention network[J]. Journal of Computer Applications, 2020, 40(3):872-877.
[5] 许夙晖,慕晓冬,赵鹏,等.利用多尺度特征与深度网络对遥感影像进行场景分类[J].测绘学报, 2016, 45(7):834-840. DOI:10.11947/j.AGCS.2016. 20150623. XU Suhui, MU Xiaodong, ZHAO Peng, et al. Scene classification of remote sensing image based on multi-scale feature and deep neural network[J]. Acta Geodaetica et Cartographica Sinica, 2016, 45(7):834-840. DOI:10.11947/j.AGCS.2016. 20150623.
[6] 蔡之灵,翁谦,叶少珍,等.基于Inception-V3模型的高分遥感影像场景分类[J].国土资源遥感, 2020, 32(3):80-89. CAI Zhiling, WENG Qian, YE Shaozhen, et al. Remote sensing image scene classification based on Inception-V3[J]. Remote Sensing for Land&Resources, 2020, 32(3):80-89.
[7] PAN S J, YANG Qiang. A survey on transfer learning[J]. IEEE Transactions on Knowledge and Data Engineering, 2010, 22(10):1345-1359.
[8] SZEGEDY C, VANHOUCKE V, IOFFE S, et al. Rethinking the inception architecture for computer vision[C]//Proceedings of 2016 IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas:IEEE, 2016:2818-2826.
[9] VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need[C]//Proceedings of the 31 st International Conference on Neural Information Processing Systems. Long Beach:ACM Press, 2017:6000-6010.
[10] SWALPA K, ANKUR D, DANFENG H, et al. Multimodal fusion transformer for remote sensing image classification[EB/OL].[2022-04-29]. https://arxiv.org/abs/2203.16952.
[11] HONG D, HAN Z, YAO J, et al. Spectral Former:rethinking hyperspectral image classification with transformers[EB/OL].[2022-04-29]. https://arxiv.org/abs/2107.02988.
[12] BAZI Y, BASHMAL L, AL RAHHAL M M, et al. Vision transformers for remote sensing image classification[J]. Remote Sensing, 2021, 13(3):516.
[13] 宋中山,梁家锐,郑禄,等.基于双向门控尺度特征融合的遥感场景分类[J].计算机应用, 2021, 41(9):2726-2735. SONG Zhongshan, LIANG Jiarui, ZHENG Lu, et al. Remote sensing scene classification based on bidirectional gated scale feature fusion[J]. Journal of Computer Applications, 2021, 41(9):2726-2735.
[14] 龚希,吴亮,谢忠,等.融合全局和局部深度特征的高分辨率遥感影像场景分类方法[J].光学学报, 2019, 39(3):0301002. GONG Xi, WU Liang, XIE Zhong, et al. Classification method of high-resolution remote sensing scenes based on fusion of global and local deep features[J]. Acta Optica Sinica, 2019, 39(3):0301002.
[15] 郑卓,方芳,刘袁缘,等.高分辨率遥感影像场景的多尺度神经网络分类法[J].测绘学报, 2018, 47(5):620-630. DOI:10.11947/j.AGCS.2018. 20170191. ZHENG Zhuo, FANG Fang, LIU Yuanyuan, et al. Joint multi-scale convolution neural network for scene classification of high resolution remote sensing imagery[J]. Acta Geodaetica et Cartographica Sinica, 2018, 47(5):620-630. DOI:10.11947/j.AGCS.2018. 20170191.
[16] HE Kaiming, ZHANG Xiangyu, REN Shaoqing, et al. Deep residual learning for image recognition[C]//Proceedings of 2016 IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas:IEEE, 2016:770-778.
[17] LIU Ze, LIN Yutong, CAO Yue, et al. Swin transformer:hierarchical vision transformer using shifted windows[C]//Proceedings of 2021 IEEE/CVF International Conference on Computer Vision. Montreal:IEEE, 2022:9992-10002.
[18] XIE Saining, GIRSHICK R, DOLLÁR P, et al. Aggregated residual transformations for deep neural networks[C]//Proceedings of 2017 IEEE Conference on Computer Vision and Pattern Recognition. Honolulu:IEEE, 2017:5987-5995.
[19] GLOROT X, BOR DE S A, BENGIO Y. Deep sparse rectifier neural networks[C]//Proceedings of the 14 th International Conference on Artificial Intelligence and Statistics. Sardinia:IEEE, 2011:315-323.
[20] WU Hang, LIU Baozhen, SU Weihua, et al. Deep filter banks for land-use scene classification[J]. IEEE Geoscience and Remote Sensing Letters, 2016, 13(12):1895-1899.
[21] ZHU Qiqi, ZHONG Yanfei, ZHAO Bei, et al. Bag-of-visual-words scene classifier with local and global features for high spatial resolution remote sensing imagery[J]. IEEE Geoscience and Remote Sensing Letters, 2016, 13(6):747-751.
[22] IOFFE S, SZEGEDY C. Batch normalization:accelerating deep network training by reducing internal covariate shift[C]//Proceedings of the 32 nd International Conference on International Conference on Machine Learning. New York:ACM Press, 2015:448-456.
[23] KINGMA D, BA J. Adam:a method for stochastic optimization[C]//Proceedings of the 3 rd International Conference for Learning Representations. Xi'an:IEEE, 2015, 1-13.
[24] ZHANG Xiangyu, ZHOU Xinyu, LIN Mengxiao, et al. ShuffleNet:an extremely efficient convolutional neural network for mobile devices[C]//Proceedings of 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Salt Lake City:IEEE, 2018:6848-6856.
[25] HU Jie, SHEN Li, ALBANIE S, et al. Squeeze-and-excitation networks[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2020, 42(8):2011-2023.
[26] SIMONYAN K, ZISSERMAN A. Very deep convolutional networks for large-scale image recognition[C]//Proceedings of 2015 International Conference for Learning Representations. San Diego:ICLR, 2015:463-476.
[27] SZEGEDY C, LIU Wei, JIA Yangqing, et al. Going deeper with convolutions[C]//Proceedings of 2015 IEEE Conference on Computer Vision and Pattern Recognition. Boston:IEEE, 2015:1-9.
[28] TAN M, LE Q V. EfficientNetV2:smaller models and faster training[C]//Proceedings of 2021 International Conference on Machine Learning. Xi'an:IEEE, 2021:10096-10106.
[29] LIU Z, MAO H, WU C Y, et al. A ConvNet for the 2020 s[EB/OL].[2022-4-29]. https://arxiv.org/abs/2201.03545.
[30] DOSOVITSKIY A, BEYER L, KOLESNIKOV A, et al. An image is worth 16 x16 words:transformers for image recognition at scale[C]//Proceedings of the 9 th International Conference on Learning Representations. Virtual Event:IEEE, 2020:1-12.
[31] CHEN Y, DAI X, CHEN D, et al. Mobile-Former:bridging mobile net and transformer[EB/OL].[2022-04-29]. https://arxiv.org/abs/2108.05895.
[32] PENG Zhiliang, HUANG Wei, GU Shanzhi, et al. Conformer:local features coupling global representations for visual recognition[C]//Proceedings of 2021 IEEE/CVF International Conference on Computer Vision. Montreal:IEEE, 2022:357-366.
[33] GUO J, HAN K, WU H, et al. CMT:convolutional neural networks meet vision transformers[EB/OL].[2022-04-29]. https://arxiv.org/abs/2107.06263.
[34] GUO M, LU C, LIU Z, et al. Visual attention network[EB/OL].[2022-04-29]. https://arxiv.org/abs/2202.09741.
[35] SELVARAJU R R, COGSWELL M, DAS A, et al. Grad-CAM:visual explanations from deep networks via gradient-based localization[J]. International Journal of Computer Vision, 2020, 128(2):336-359.
[36] SPRINGENBERG J T, DOSOVITSKIY A, BROX T, et al. Striving for simplicity:the all convolutional net[EB/OL].[2022-04-29]. https://arxiv.org/abs/1412.6806 v1

面向遥感图像场景分类的GLFFNet模型

GLFFNet model for remote sensing image scene classification

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

[1]	安晓亚, 朱余德, 晏雄锋. 卷积神经网络支持下的建筑物选取方法[J]. 测绘学报, 2023, 52(9): 1574-1583.
[2]	何佳星, 郑南山, 丁锐, 张克非, 陈天悦. 粒子群优化卷积神经网络GNSS-IR土壤湿度反演方法[J]. 测绘学报, 2023, 52(8): 1286-1297.
[3]	连仁包, 张振敏, 廖一鹏, 邹长忠, 黄立勤. 结合测地距离场与曲线平滑的遥感图像道路中心线快速提取[J]. 测绘学报, 2023, 52(8): 1317-1329.
[4]	张艺超, 郑向涛, 卢孝强. 基于层级Transformer的高光谱图像分类方法[J]. 测绘学报, 2023, 52(7): 1139-1147.
[5]	张蒙蒙, 李伟, 刘欢, 赵旭东, 陶然. 基于形态变换与空间逻辑聚合的高光谱森林树种分类[J]. 测绘学报, 2023, 52(7): 1202-1211.
[6]	顾小虎, 李正军, 缪健豪, 李星华, 沈焕锋. 高分遥感影像双通道并行混合卷积分类方法[J]. 测绘学报, 2023, 52(5): 798-807.
[7]	赵冰冰, 谭骁勇, 杨学习, 石岩, 邓敏. 融合地理条件驱动效应和图卷积的土地利用演化模拟CA模型[J]. 测绘学报, 2023, 52(5): 831-842.
[8]	余东行, 徐青, 赵传, 郭海涛, 卢俊, 林雨准, 刘相云. 注意力引导特征融合与联合学习的遥感影像场景分类[J]. 测绘学报, 2023, 52(4): 624-637.
[9]	刘帅, 李笑迎, 于梦, 邢光龙. 高分辨率遥感图像双解耦语义分割网络模型[J]. 测绘学报, 2023, 52(4): 638-647.
[10]	吴晨昊, 向隆刚, 张叶廷, 吴华意. 基于地理空间感知型表征学习的轨迹相似度计算[J]. 测绘学报, 2023, 52(4): 670-678.
[11]	张永显, 马国锐, 崔志祥, 张志军. 面向大视角差的无人机影像序列学习型特征匹配[J]. 测绘学报, 2023, 52(2): 230-243.
[12]	孙根云, 王鑫, 安娜, 张爱竹. 基于多源遥感的大尺度高分辨率不透水面深度学习提取方法[J]. 测绘学报, 2023, 52(2): 272-282.
[13]	张广斌, 高贤君, 冉树浩, 杨元维, 李丽珊, 张妍. 高分遥感影像云雪共存区轻量云高精度检测方法[J]. 测绘学报, 2023, 52(1): 93-107.
[14]	杨敏, 陈果, 李连营, 黄浩然, 苗静, 晏雄锋. 序列卷积神经网络支持下线状地图目标的分段方法[J]. 测绘学报, 2023, 52(1): 108-116.
[15]	白坤, 慕晓冬, 陈雪冰, 朱永清, 尤轩昂. 融合半监督学习的无监督遥感影像场景分类[J]. 测绘学报, 2022, 51(5): 691-702.