[1]王瑞婷,王海燕,陈晓,等.基于混合卷积与三重注意力的高光谱图像分类网络[J].智能系统学报,2023,18(2):260-269.[doi:10.11992/tis.202204002]
WANG Ruiting,WANG Haiyan,CHEN Xiao,et al.Hyperspectral image classification based on hybrid convolutional neural network with triplet attention[J].CAAI Transactions on Intelligent Systems,2023,18(2):260-269.[doi:10.11992/tis.202204002]
点击复制
WANG Ruiting,WANG Haiyan,CHEN Xiao,et al.Hyperspectral image classification based on hybrid convolutional neural network with triplet attention[J].CAAI Transactions on Intelligent Systems,2023,18(2):260-269.[doi:10.11992/tis.202204002]
基于混合卷积与三重注意力的高光谱图像分类网络
《智能系统学报》[ISSN 1673-4785/CN 23-1538/TP] 卷:
18
期数:
2023年第2期
页码:
260-269
栏目:
学术论文—机器学习
出版日期:
2023-05-05
- Title:
- Hyperspectral image classification based on hybrid convolutional neural network with triplet attention
- Keywords:
- remote sensing; hyperspectral image classification; deep learning; feature extraction; dimension reduction; depth-separable convolution; attention mechanism; hybrid convolutional neural network
- 分类号:
- TP751
- 摘要:
- 针对高光谱图像光谱维度高、现有网络无法提供深度级的多层次特征，从而影响分类精度和速度的问题。首先采用核主成分分析对高光谱图像进行降维，使降维后的数据具有最佳区分度，提出了一种基于混合卷积与三重注意力的卷积神经网络(hybrid convolutional neural network with triplet attention, HCTA-Net)模型，该模型设计了一种基于三维、二维和一维卷积的混合卷积神经网络，通过不同维度卷积神经网络的融合，提取高光谱图像精细的光谱–空间联合特征。在二维卷积中加入深度可分离卷积，减少了模型参数，同时引入三重注意力机制，使用三分支结构实现跨维度信息交互，抑制无用的特征信息。在Indian Pines、Salinas和Pavia University数据集上的实验结果表明，本文提出的模型优于其他对比方法，总体分类精度分别达到了99.16%、99.87%和99.76%。
- Abstract:
- To solve the problems of a high spectral dimension of hyperspectral images and the failure of the existing network to provide multilevel features at the depth level, which affects the classification accuracy and speed, the kernel principal component analysis is used to reduce the dimensionality of hyperspectral images to have the best data differentiation after dimensionality reduction, and a hybrid convolutional neural network with triplet attention (HCTA-Net) model is proposed to design a hybrid model based on 3D, 2D, and 1D CNN to extract the fine spectral–spatial joint features through the fusion of different dimension convolutions. The model also adds depthwise separable convolution into the 2D-CNN to reduce the model parameters and simultaneously introduces a triplet attention mechanism, which uses a three-branch structure to achieve cross-dimensional information interaction to inhibit useless feature information. Experimental results on the Indian Pines, Salinas, and Pavia University datasets show that the proposed model is superior to other comparison methods, and the overall classification accuracy reaches 99.16%, 99.87, and 99.76%, respectively.
- 参考文献/References:
-
[1] 王锦锦, 李真, 朱玉玲. 高光谱影像在海洋环境监测中的应用[J]. 卫星应用, 2019(8): 36–40
[2] 高暖. 高光谱图像地物及目标识别[D]. 西安: 西安电子科技大学, 2019: 1?5.
GAO Nuan. Terrain and target detection in hyperspectral remote sensing imagery[D]. Xi’an: Xidian University, 2019: 1?5.
[3] HUGHES G. On the mean accuracy of statistical pattern recognizers[J]. IEEE transactions on information theory, 1968, 14(1): 55–63.
[4] BLANZIERI E, MELGANI F. Nearest neighbor classification of remote sensing images with the maximal margin principle[J]. IEEE transactions on geoscience and remote sensing, 2008, 46(6): 1804–1811.
[5] TARABALKA Y, FAUVEL M, CHANUSSOT J, et al. SVM- and MRF-based method for accurate classification of hyperspectral images[J]. IEEE geoscience and remote sensing letters, 2010, 7(4): 736–740.
[6] 陈善学, 何宇峰. 基于空谱字典的加权联合稀疏表示高光谱图像分类[J]. 光学学报, 2023, 43(1): 0110002–1-11
CHEN Shanxue, HE Yufeng. Weighted joint sparse representation hyperspectral image classification based on spatial-spectral diction[J]. Acta optica sinica, 2023, 43(1): 0110002–1-11
[7] 熊余, 单德明, 姚玉, 等. 多特征融合下的高光谱图像混合卷积分类[J]. 红外技术, 2022, 44(1): 9–20
XIONG Yu, SHAN Deming, YAO Yu, et al. Hyperspectral image hybrid convolution classification under multi-feature fusion[J]. Infrared technology, 2022, 44(1): 9–20
[8] FAUVEL M, TARABALKA Y, BENEDIKTSSON J A, et al. Advances in spectral-spatial classification of hyperspectral images[J]. Proceedings of the IEEE, 2013, 101(3): 652–675.
[9] ZHONG Ping, GONG Zhiqiang. A hybrid DBN and CRF model for spectral-spatial classification of hyperspectral images[J]. Statistics, optimization & information computing, 2017, 5(2): 75–98.
[10] LIU Jinxiang, ZHANG Kefei, WU Suqin, et al. An investigation of a multidimensional CNN combined with an attention mechanism model to resolve small-sample problems in hyperspectral image classification[J]. Remote sensing, 2022, 14(3): 785.
[11] HU Wei, HUANG Yangyu, WEI Li, et al. Deep convolutional neural networks for hyperspectral image classification[J]. Journal of sensors, 2015, 2015: 258619.
[12] XU Xiaodong, LI Wei, RAN Qiong, et al. Multisource remote sensing data classification based on convolutional neural network[J]. IEEE transactions on geoscience and remote sensing, 2018, 56(2): 937–949.
[13] ZHANG Xin, WANG Yongcheng, ZHANG Ning, et al. SSDANet: spectral-spatial three-dimensional convolutional neural network for hyperspectral image classification[J]. IEEE access, 2020, 8: 127167–127180.
[14] GUO Hao, LIU Jianjun, YANG Jinlong, et al. Deep collaborative attention network for hyperspectral image classification by combining 2-D CNN and 3-D CNN[J]. IEEE journal of selected topics in applied earth observations and remote sensing, 2020, 13: 4789?4802.
[15] HU Jie, SHEN Li, SUN Gang. Squeeze-and-excitation networks[C]//IEEE/CVF Conference on Computer Vision and Pattern Recognition. Salt Lake City: IEEE, 2018: 7132?7141.
[16] WOO S, PARK J, LEE J Y, et al. CBAM: convolutional block attention module[M]//Computer Vision-ECCV 2018. Cham: Springer International Publishing, 2018: 3?19.
[17] FANG Bei, LI Ying, ZHANG Haokui, et al. Hyperspectral images classification based on dense convolutional networks with spectral-wise attention mechanism[J]. Remote sensing, 2019, 11(2): 159.
[18] ROY S K, MANNA S, SONG Tiecheng, et al. Attention-based adaptive spectral-spatial kernel ResNet for hyperspectral image classification[J]. IEEE transactions on geoscience and remote sensing, 2021, 59(9): 7831–7843.
[19] AHMAD M, KHAN A M, MAZZARA M, et al. Attention mechanism meets with hybrid dense network for hyperspectral image classification[EB/OL].(2022?01?04)[2022?12?12].https://arxiv.org/abs/2201.01001.
[20] CHOLLET F. Xception: deep learning with depthwise separable convolutions[C]//IEEE Conference on Computer Vision and Pattern Recognition. Honolulu: IEEE, 2017: 1800?1807.
[21] CUI Ying, YU Zikun, HAN Jiacheng, et al. Dual-triple attention network for hyperspectral image classification using limited training samples[J]. IEEE geoscience and remote sensing letters, 2022, 19: 1–5.
[22] 李昌元, 刘国栋, 谭博. 基于PCA和KPCA的高光谱遥感数据降维对比研究[J]. 地理空间信息, 2022, 20(7): 89–93,103
LI Changyuan, LIU Guodong, TAN Bo. Comparative study on hyperspectral remote sensing data dimensionality reduction based on PCA and KPCA[J]. Geospatial information, 2022, 20(7): 89–93,103
[23] MELGANI F, BRUZZONE L. Classification of hyperspectral remote sensing images with support vector machines[J]. IEEE transactions on geoscience and remote sensing, 2004, 42(8): 1778–1790.
[24] CHEN Yushi, JIANG Hanlu, LI Chunyang, et al. Deep feature extraction and classification of hyperspectral images based on convolutional neural networks[J]. IEEE transactions on geoscience and remote sensing, 2016, 54(10): 6232–6251.
[25] ZHONG Zilong, LI J, LUO Zhiming, et al. Spectral-spatial residual network for hyperspectral image classification: a 3-D deep learning framework[J]. IEEE transactions on geoscience and remote sensing, 2018, 56(2): 847–858.
[26] ROY S K, KRISHNA G, DUBEY S R, et al. HybridSN: exploring 3-D–2-D CNN feature hierarchy for hyperspectral image classification[J]. IEEE geoscience and remote sensing letters, 2020, 17(2): 277–281.
[27] 王爱丽, 刘美红, 薛冬, 等. 结合动态卷积和三重注意力机制的高光谱图像分类[J]. 激光与光电子学进展, 2022, 59(10): 341–351
WANG Aili, LIU Meihong, XUE Dong, et al. Hyperspectral image classification combined dynamic convolution with triplet attention mechanism[J]. Laser & optoelectronics progress, 2022, 59(10): 341–351
[28] ZHANG Yangming, YANG Kun, YUAN Lei. A hyperspectral image classification method with CNN based on attention-enhanced spectral and spatial features[J]. Journal of physics:conference series, 2021, 2006(1): 012033.
- 相似文献/References:
-
[1]陈立伟,房赫,朱海峰.多视图主动学习的多样性样本选择方法研究[J].智能系统学报,2021,16(6):1007.[doi:10.11992/tis.202007037]
CHEN Liwei,FANG He,ZHU Haifeng.Diversity sample selection method of multiview active learning classification[J].CAAI Transactions on Intelligent Systems,2021,16():1007.[doi:10.11992/tis.202007037]
备注/Memo
收稿日期:2022-04-02。
基金项目:国家自然科学基金重点项目（62031021）.
作者简介:王瑞婷,硕士研究生,主要研究方向为复杂环境智能信息感知;王海燕,教授,博士生导师,主要研究方向为智能感知技术。负责国家自然基金、国家科技重大专项、国家863计划等项目20余项,发表学术论文80余篇;陈晓,讲师,主要研究方向为多目标跟踪理论和AI技术
通讯作者:王海燕. E-mail：hywang@sust.edu.cn
更新日期/Last Update:
1900-01-01