浏览全部资源
扫码关注微信
Published： 16 November 2023 ，
移动端阅览
杜艳玲， 吴天宇， 陈括， 陈刚， 宋巍. 2023. 融合上下文和注意力的海洋涡旋小目标检测. 中国图象图形学报， 28(11):3509-3519
Du Yanling， Wu Tianyu， Chen Kuo， Chen Gang， Song Wei. 2023. Small object detection for ocean eddies using contextual information and attention mechanism. Journal of Image and Graphics， 28(11):3509-3519
目的
2
海洋涡旋精准检测是揭示海洋涡旋演变规律及其与其他海洋现象相互作用的基础。然而，海洋涡旋在其活跃海域呈现小尺度目标、密集分布的特点，导致显著的检测精度低问题。传统方法受限于人工设计参数缺乏泛化能力，而深度学习模型的高采样率在检测小目标过程中底层细节和轮廓等信息损失严重，使得目标检测轮廓与目标真实轮廓相差甚远。针对海洋涡旋小目标特点导致检测精度低，高采样率深度模型检测轮廓不精确的问题，提出一种改进的U-Net网络。
方法
2
该模型基于渐进式采样结构，为获取上下文信息提升不同极性海洋涡旋目标的检测精度，增加上下文特征融合模块；为增加该模块对海洋涡旋小目标的关注，在特征融合前对最底层特征嵌入残差注意力模块，使模型可以更多关注海洋涡旋的轮廓信息。最后引入数据扩充方法缓解模型存在的过拟合问题。
结果
2
本文以南大西洋的卫星海表面高度数据集开展实验，结果表明，本文模型检测准确率达到了93.24%，同时在海洋涡旋的检测数量上与真实结果更加接近，验证了模型在小目标检测方面的性能更加优秀。
结论
2
本文提出的海洋涡旋小目标检测模型，在检测海洋涡旋的性能与海洋涡旋目标轮廓精准度方面均显著优于全卷积神经网络（fully convolutional network，FCN）等深度学习模型。
Objective
2
Ocean eddies are responsible for most of the material transportation and energy transfers in the ocean. The accurate detection of these eddies serves as the basis for revealing the evolution of ocean eddies and their interactions with other marine phenomena. However， small-scale objects and dense distribution are often observed in the active area of ocean eddies， which leads to problem of low detection accuracy. Traditional detection methods are limited by the poor generalizability of the artificial parameter design. These methods also have poor ocean eddy detection accuracy compared with deep learning methods. However， a deep learning model with high sampling rate loses the underlying details and contour information in the process of small target detection. The target detection contour is located far from the real contour of the target. To address the low detection accuracy caused by the loss of low-level detail information and contour information of small-scale ocean eddy targets， this paper proposes an improved U-Net network.
Method
2
Based on the U-shaped progressive sampling network， a context feature fusion module is added to fuse the features of each coding layer， and a residual attention mechanism is added to the target features before the feature fusion in order for the model to pay attention to the contour information of the ocean eddies. A data augmentation method is then introduced to reduce the overfitting problem of the model. Feature fusion is carried out through the context feature fusion module， which takes the three-layer feature map of the U-shaped structure coding layer of the U-Net network as input， the lowest-level feature map as the target feature， and the last two-layer feature map as the context and target features. The context feature map is initially upsampled to the same size as the lowest-level feature through the deconvolution structure， and the number of channels is reduced to 1/2 of the lowest-level feature in order to prevent the amount of information of the context feature from exceeding that of the target feature. L2 norm and ReLU are then used to achieve the fusion of context and target features. The proposed model uses two contextual feature fusion modules， which take the first to third layer feature maps of the encoding layer as input and the second to fourth layer feature maps as input， respectively. The residual attention mechanism consists of two processing channels. The first channel has a residual structure （batch norm， conv of 1 × 1 kernel and multiple concatenation of ReLU） that prevents gradient disappearance and extracts certain contour information， while the second channel comprises a down-up sampling layer and a sigmoid layer to extract high-level semantic information. To effectively reduce the over-fitting phenomenon， random region sampling and random mask processing are used for data augmentation. In the experiment， the model is trained in the NVIDIA GTX 1080Ti GPU environment， where its initial learning rate is set to 1 × 10
-3
， the loss function is optimized by the Adam optimizer， the batch size of the model training is set to 16， and the number of iterations is set to 200.
Result
2
The satellite sea surface height dataset of the South Atlantic is used for the experiments. Ablation experiments are carried out to test the influence of each module on the performance of the ocean eddies detection model. The effects of adding the context feature fusion module， adding the attention mechanism module， and adding both modules at the same time are compared， and the detection effect after adding the data augmentation method is analyzed. In the ablation experiment， due to the introduction of the contextual feature fusion module and the residual attention mechanism， the model can fuse the contextual features of the ocean eddies in different feature layers， and the network can extract additional low-level spatial details of the ocean eddies. Each module improves the detection performance of the model， and the optimal detection accuracy of the model after using the data augmentation method reaches 93.27%. Compared with other deep learning models， the proposed model has a detection accuracy of up to 93.24%， and its detected number of ocean eddies is closer to the truth， thereby verifying its excellent performance in small target detection. Meanwhile， compared with the fully convolutional network （FCN） model， the proposed model can detect more small-scale ocean eddy targets， and the detected ocean eddy target contour is closer to the truth， thereby verifying the positive effect of progressive sampling on small target detection.
Conclusion
2
The proposed model significantly outperforms the other deep learning models in detecting ocean eddies. Compared with the state-of-the-art， the proposed model achieves a higher small target detection accuracy， and the detected contour of ocean eddies is closer to the truth.
海洋涡旋小目标检测语义分割注意力机制特征融合
ocean eddysmall object detectionsemantic segmentationattention mechanismsfeature fusion
Ali Sadarjoen I and Post F H. 2000. Detection， quantification， and tracking of vortices using streamline geometry. Computers and Graphics， 24（3）： 333-341 ［DOI： 10.1016/S0097-8493（00）00029-7http://dx.doi.org/10.1016/S0097-8493（00）00029-7］
Cao Y， Xu J R， Lin S， Wei F Y and Hu H. 2019. Gcnet： non-local networks meet squeeze-excitation networks and beyond//Proceedings of 2019 IEEE/CVF International Conference on Computer Vision Workshops. Seoul， Korea （South）： IEEE： 1971-1980 ［DOI： 10.1109/ICCVW.2019.00246http://dx.doi.org/10.1109/ICCVW.2019.00246］
Chaigneau A， Gizolme A and Grados C. 2008. Mesoscale eddies off Peru in altimeter records： identification algorithms and eddy spatio-temporal patterns. Progress in Oceanography， 79（2-4）： 106-119 ［DOI： 10.1016/j.pocean.2008.10.013http://dx.doi.org/10.1016/j.pocean.2008.10.013］
Chelton D B， Schlax M G， Samelson R M and De Szoeke R A. 2007. Global observations of large oceanic eddies. Geophysical Research Letters， 34（15）： #L15606 ［DOI： 10.1029/2007GL030812http://dx.doi.org/10.1029/2007GL030812］
Chelton D B， Schlax M G and Samelson R M. 2011. Global observations of nonlinear mesoscale eddies. Progress in Oceanography， 91（2）： 167-216 ［DOI： 10.1016/j.pocean.2011.01.002http://dx.doi.org/10.1016/j.pocean.2011.01.002］
Chen G X， Hou Y J and Chu X Q. 2011. Mesoscale eddies in the South China Sea： mean properties， spatiotemporal variability， and impact on thermohaline structure. Journal of Geophysical Research： Oceans， 116（C6）： #C06018 ［DOI： 10.1029/2010jc006716http://dx.doi.org/10.1029/2010jc006716］
Chen L C， Papandreou G， Schroff F and Adam H. 2017. Rethinking atrous convolution for semantic image segmentation ［EB/OL］. ［2022-10-10］. https://arxiv.org/pdf/1706.05587.pdfhttps://arxiv.org/pdf/1706.05587.pdf
Chen L C， Zhu Y K， Papandreou G， Schroff F and Adam H. 2018. Encoder-decoder with atrous separable convolution for semantic image segmentation//Proceedings of the 15th European Conference on Computer Vision. Berlin， Germany： Springer： 833-851 ［DOI： 10.1007/978-3-030-01234-2_49http://dx.doi.org/10.1007/978-3-030-01234-2_49］
Doglioli A M， Blanke B， Speich S and Lapeyre G. 2007. Tracking coherent structures in a regional ocean model with wavelet analysis： application to Cape Basin eddies. Journal of Geophysical Research： Oceans， 112（C5）： #C05043 ［DOI： 10.1029/2006JC003952http://dx.doi.org/10.1029/2006JC003952］
Dong Z Y， Du Z H， Wu S S， Li Y D， Zhang F， Liu R Y. 2022. An automatic marine mesoscale eddy detection model based on improved U-Net network. Acta Oceanologica Sinica， 44（2）： 123-131
董子意， 杜震洪， 吴森森， 李亚东， 张丰， 刘仁义. 2022. 基于改进U-Net网络的海洋中尺度涡自动检测模型. 海洋学报， 44（2）： 123-131 ［DOI： 10.12284/hyxb2022038http://dx.doi.org/10.12284/hyxb2022038］
Du Y L， Liu Q Q， Wang L L， Xu X， Wei Q M and Song W. 2022. Multi-scale rotating anchor mechanism based automatic detection of ocean mesoscale eddy. Journal of Image and Graphics， 27（10）： 3092-3101
杜艳玲， 刘倩倩， 王丽丽， 徐鑫， 魏泉苗， 宋巍. 2022. 融合多尺度旋转锚机制的海洋中尺度涡自动检测. 中国图象图形学报， 27（10）： 3092-3101 ［DOI： 10.11834/jig.210286http://dx.doi.org/10.11834/jig.210286］
Faghmous J H， Le M， Uluyol M， Kumar V and Chatterjee S. 2013. A parameter-free spatio-temporal pattern mining model to catalog global ocean dynamics//Proceedings of the 13th IEEE International Conference on Data Mining. Dallas， USA： IEEE： 151-160 ［DOI： 10.1109/ICDM.2013.162http://dx.doi.org/10.1109/ICDM.2013.162］
Faghmous J H， Styles L， Mithal V， Boriah S， Liess S， Kumar V， Vikebø F and Dos Santos Mesquita M. 2012. EddyScan： a physically consistent ocean eddy monitoring application//Proceedings of 2012 Conference on Intelligent Data Understanding. Boulder， USA： IEEE： 96-103 ［DOI： 10.1109/CIDU.2012.6382189http://dx.doi.org/10.1109/CIDU.2012.6382189］
Fan Z L， Zhong G Q， Wei H X and Li H T. 2020. EDNet： a mesoscale eddy detection network with multi-modal data//Proceedings of 2020 International Joint Conference on Neural Networks. Glasgow， UK： IEEE： 1-7 ［DOI： 10.1109/IJCNN48605.2020.9206613http://dx.doi.org/10.1109/IJCNN48605.2020.9206613］
Fu C Y， Liu W， Ranga A， Tyagi A and Berg A C. 2017. DSSD： deconvolutional single shot detector ［EB/OL］. ［2022-10-02］. https://arxiv.org/pdf/1701.06659.pdfhttps://arxiv.org/pdf/1701.06659.pdf
Fu J， Liu J， Tian H J， Li Y， Bao Y J， Fang Z W and Lu H Q. 2019. Dual attention network for scene segmentation//Proceedings of 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Long Beach， USA： IEEE： 3141-3149 ［DOI： 10.1109/CVPR.2019.00326http://dx.doi.org/10.1109/CVPR.2019.00326］
He K M， Gkioxari G， Dollar P and Girshick R. 2017. Mask R-CNN//Proceedings of 2017 IEEE International Conference on Computer Vision. Venice， Italy： IEEE： 2980-2988 ［DOI： 10.1109/ICCV.2017.322http://dx.doi.org/10.1109/ICCV.2017.322］
Henson S A and Thomas A C. 2008. A census of oceanic anticyclonic eddies in the Gulf of Alaska. Deep Sea Research Part I： Oceanographic Research Papers， 55（2）： 163-176 ［DOI： 10.1016/j.dsr.2007.11.005http://dx.doi.org/10.1016/j.dsr.2007.11.005］
Jeong J， Park H and Kwak N. 2017. Enhancement of SSD by concatenating feature maps for object detection ［EB/OL］. ［2022-10-02］. https://arxiv.org/pdf/1705.09587.pdfhttps://arxiv.org/pdf/1705.09587.pdf
Jia K X， Ma Z H， Zhu R and Li Y G. 2022. Attention-mechanism-based light single shot multiBox detector modelling improvement for small object detection on the sea surface. Journal of Image and Graphics， 27（4）： 1161-1175
贾可心， 马正华， 朱蓉， 李永刚. 2022. 注意力机制改进轻量SSD模型的海面小目标检测. 中国图象图形学报， 27（4）： 1161-1175 ［DOI： 10.11834/jig.200517http://dx.doi.org/10.11834/jig.200517］
Lguensat R， Sun M， Fablet R， Tandeo P， Mason E and Chen G. 2018. EddyNet： a deep neural network for pixel-wise classification of oceanic eddies//Proceedings of IGARSS 2018 IEEE International Geoscience and Remote Sensing Symposium. Valencia， Spain： IEEE： 1764-1767 ［DOI： 10.1109/IGARSS.2018.8518411http://dx.doi.org/10.1109/IGARSS.2018.8518411］
Li X， Zhong Z S， Wu J L， Yang Y B， Lin Z C and Liu H. 2019. Expectation-maximization attention networks for semantic segmentation//Proceedings of 2019 IEEE/CVF International Conference on Computer Vision. Seoul， Korea （South）： IEEE： 9166-9175 ［DOI： 10.1109/ICCV.2019.00926http://dx.doi.org/10.1109/ICCV.2019.00926］
Lim J S， Astrid M， Yoon H J and Lee S I. 2021. Small object detection using context and attention//Proceedings of 2021 International Conference on Artificial Intelligence in Information and Communication. Jeju Island， Korea （South）： IEEE： 181-186 ［DOI： 10.1109/ICAIIC51459.2021.9415217http://dx.doi.org/10.1109/ICAIIC51459.2021.9415217］
Lin T Y， Dollar P， Girshick R， He K M， Hariharan B and Belongie S. 2017. Feature pyramid networks for object detection//Proceedings of 2017 IEEE Conference on Computer Vision and Pattern Recognition （CVPR）. Honolulu， USA： IEEE： 936-944 ［DOI： 10.1109/CVPR.2017.106http://dx.doi.org/10.1109/CVPR.2017.106］
Liu Z M， Gao G Y， Sun L and Fang Z Y. 2021. HRDNet： high-resolution detection network for small objects//Proceedings of 2021 IEEE International Conference on Multimedia and Expo. Shenzhen， China： IEEE： 1-6 ［DOI： 10.1109/ICME51207.2021.9428241http://dx.doi.org/10.1109/ICME51207.2021.9428241］
Long J， Shelhamer E and Darrell T. 2015. Fully convolutional networks for semantic segmentation//Proceedings of 2015 IEEE Conference on Computer Vision and Pattern Recognition. Boston， USA： IEEE： 3431-3440 ［DOI： 10.1109/CVPR.2015.7298965http://dx.doi.org/10.1109/CVPR.2015.7298965］
Mason E， Pascual A and McWilliams J C. 2014. A new sea surface height-based code for oceanic mesoscale eddy tracking. Journal of Atmospheric and Oceanic Technology， 31（5）： 1181-1188 ［DOI： 10.1175/JTECH-D-14-00019.1http://dx.doi.org/10.1175/JTECH-D-14-00019.1］
McWilliams J C. 1984. The emergence of isolated， coherent vortices in turbulent flow. AIP Conference Proceedings， 106（1）： 205-221 ［DOI： 10.1063/1.34273http://dx.doi.org/10.1063/1.34273］
Nencioli F， Dong C M， Dickey T， Washburn L and McWilliams J C. 2010. A vector geometry–based eddy detection algorithm and its application to a high-resolution numerical model product and high-frequency radar surface velocities in the Southern California Bight. Journal of Atmospheric and Oceanic Technology， 27（3）： 564-579 ［DOI： 10.1175/2009JTECHO725.1http://dx.doi.org/10.1175/2009JTECHO725.1］
Ronneberger O， Fischer P and Brox T. 2015. U-Net： convolutional networks for biomedical image segmentation//Proceedings of the 18th International Conference on Medical Image Computing and Computer-Assisted Intervention. Munich， Germany： Springer： 234-241 ［DOI： 10.1007/978-3-319-24574-4_28http://dx.doi.org/10.1007/978-3-319-24574-4_28］
Shen B， Chen Y， Yang C and Liu B W. 2020. Computer vision detection and analysis of mesoscale eddies in marine science. Frontiers of Data and Domputing， 2（6）： 30-41
沈飙， 陈扬， 杨琛， 刘博文. 2020. 海洋科学中尺度涡的计算机视觉检测和分析方法. 数据与计算发展前沿， 2（6）： 30-41 ［DOI： 10.11871/jfdc.issn.2096-742X.2020.06.004http://dx.doi.org/10.11871/jfdc.issn.2096-742X.2020.06.004］
Wang F， Jiang M Q， Qian C， Yang S， Li C， Zhang H G， Wang X G and Tang X O. 2017. Residual attention network for image classification//Proceedings of 2017 IEEE Conference on Computer Vision and Pattern Recognition. Honolulu， USA： IEEE： 6450-6458 ［DOI： 10.1109/CVPR.2017.683http://dx.doi.org/10.1109/CVPR.2017.683］
Williams S， Hecht M， Petersen M， Strelitz R， Maltrud M， Ahrens J， Hlawitschka M and Hamann B. 2011. Visualization and analysis of eddies in a global ocean simulation. Computer Graphics Forum， 30（3）： 991-1000 ［DOI： 10.1111/j.1467-8659.2011.01948.xhttp://dx.doi.org/10.1111/j.1467-8659.2011.01948.x］
Woo S， Park J， Lee J Y and Kweon I S. 2018. CBAM： convolutional block attention module//Proceedings of the 15th European Conference on Computer Vision. Munich， Germany： Springer： 3-19 ［DOI： 10.1007/978-3-030-01234-2_1http://dx.doi.org/10.1007/978-3-030-01234-2_1］
Xu G J， Cheng C， Yang W X， Xie W H， Kong L M， Hang R L， Ma F R， Dong C M and Yang J S. 2019. Oceanic eddy identification using an AI scheme. Remote Sensing， 11（11）： #1349 ［DOI： 10.3390/rs11111349http://dx.doi.org/10.3390/rs11111349］
Zhao H S， Shi J P， Qi X J， Wang X G and Jia J Y. 2017. Pyramid scene parsing network//Proceedings of 2017 IEEE Conference on Computer Vision and Pattern Recognition. Honolulu， USA： IEEE： 6230-6239 ［DOI： 10.1109/CVPR.2017.660http://dx.doi.org/10.1109/CVPR.2017.660］
相关作者
相关机构
京公网安备11010802024621