Seafloor classification based on combined multibeam bathymetry and backscatter using deep convolution neural network

doi:10.11947/j.AGCS.2021.20200065

Abstract

Abstract: Seafloor classification is of great significance for the development and utilization of marine resources and marine scientific research. At present, multibeam detection is one of the effective methods to achieve large-scale seafloor classification. Seafloor classification is usually based on the angular response (AR) features and backscatter image features extracted by using multibeam backscatter. Because the feature source is relatively single and classifier structure is simple, the classification accuracy is often not high. This paper proposes a seafloor classification method based on convolutional neural networks (CNN). In addition to backscatter features, bathymetry features are also used to classify. The feature vectors are converted into waveform maps, and then input to the convolutional neural network for training and classification. The experiment compares different feature combination models and four conventional classifiers: BP network, support vector machine (SVM), k-nearest neighbor (KNN), and random forest (RF). The overall classification accuracy of the experiment reaches 94.86%, the kappa coefficient up to 0.93, and it takes 1 min 25 s. The accuracy has obvious advantages and the efficiency is relatively high. This method can effectively obtain the seafloor information in two different data types, give full play to the characteristics of convolutional neural network weight sharing, high efficiency, and achieve high-resolution seafloor classification. This paper provides a reference for the seafloor classification based on multibeam.

Key words: multibeam, backscatter image, angular response, seafloor classification, convolutional neural network

CLC Number:

P229.1

YANG Fanlin, ZHU Zhengren, LI Jiabiao, FENG Chengkai, XING Zhe, WU Ziyin. Seafloor classification based on combined multibeam bathymetry and backscatter using deep convolution neural network[J]. Acta Geodaetica et Cartographica Sinica, 2021, 50(1): 71-84.

References

[1] 何林帮. 基于多波束和浅剖的海底浅表层沉积物分类关键问题研究[J]. 测绘学报, 2016, 45(12):1498-1512. DOI:10.11947/j.AGCS.2016.20160466. HE Linbang. Research on key issues of sediment classification for seabed and sub-bottom based on multi-beam and sub-bottom profile echo intensity[J]. Acta Geodaetica et Cartographica Sinica, 2016, 45(12):1498-1512. DOI:10.11947/j.AGCS.2016.20160466.
[2] HERKVL K, PETERSON A, PAEKIVI S. Applying multibeam sonar and mathematical modeling for mapping seabed substrate and biota of offshore shallows[J]. Estuarine, Coastal and Shelf Science, 2017, 192:57-71.
[3] 唐秋华, 纪雪, 丁继胜, 等. 多波束声学底质分类研究进展与展望[J]. 海洋科学进展, 2019, 37(1):1-10. TANG Qiuhua, JI Xue, DING Jisheng, et al. Research progress and prospect of acoustic seabed classification using multibeam echo sounder[J]. Advances in Marine Science, 2019, 37(1):1-10.
[4] ZHAO Dineng, WU Ziyin, ZHOU Jieqiong, et al. A new method of automatic SVP optimization based on MOV algorithm[J]. Marine Geodesy, 2015, 38(3):225-240.
[5] ZHOU Jieqiong, WU Ziyin, JIN Xianglong, et al. Observations and analysis of giant sand wave fields on the Taiwan Banks, northern South China Sea[J]. Marine Geology, 2018, 406:132-141.
[6] TEGOWSKI J. Acoustical classification of the bottom sediments in the southern Baltic Sea[J]. Quaternary International, 2005, 130(1):153-161.
[7] PRESTON J. Automated acoustic seabed classification of multibeam images of Stanton Banks[J]. Applied Acoustics, 2009, 70(10):1277-1287.
[8] 熊明宽, 吴自银, 李守军, 等. 基于遗传小波神经网络的海底声学底质识别分类[J]. 海洋学报, 2014, 36(5):90-97. XIONG Mingkuan, WU Ziyin, LI Shoujun, et al. Wavelet neural network identification and classification of sediment seabed sonar images based on genetic algorithms[J]. Acta Oceanologica Sinica, 2014, 36(5):90-97.
[9] HAMILTON L J, PARNUM I. Acoustic seabed segmentation from direct statistical clustering of entire multibeam sonar backscatter curves[J]. Continental Shelf Research, 2011, 31(2):138-148.
[10] HASAN R C, IERODIACONOU D, LAURENSON L, et al. Integrating multibeam backscatter angular response, mosaic and bathymetry data for benthic habitat mapping[J]. PLoS One, 2014, 9(5):e97339.
[11] MONTEYS X, HUNG P, SCOTT G, et al. The use of multibeam backscatter angular response for marine sediment characterisation by comparison with shallow electromagnetic conductivity[J]. Applied Acoustics, 2016, 112:181-191.
[12] 金绍华, 李家彪, 吴自银, 等. 海底底质分类反向散射强度三维概率密度法[J]. 测绘学报, 2019, 48(1):124-131. DOI:10.11947/j.AGCS.2019.20170631. JIN Shaohua, LI Jiabiao, WU Ziyin, et al. 3D histogram of backscatter strength for seafloor substrates classification[J]. Acta Geodaetica et Cartographica Sinica, 2019, 48(1):124-131. DOI:10.11947/j.AGCS.2019.20170631.
[13] 吕海龙, 杜德文, 石学法, 等. 基于测深数据的胶州湾底质类型估计方法[J]. 海洋科学进展, 2004, 22(3):328-333. LV Hailong, DU Dewen, SHI Xuefa, et al. Estimation method of bottom sediment type in the Jiaozhou Bay based on the bathymetric data[J]. Advances in Marine Science, 2004, 22(3):328-333.
[14] MARSH I, BROWN C. Neural network classification of multibeam backscatter and bathymetry data from Stanton Bank (Area IV)[J]. Applied Acoustics, 2009, 70(10):1269-1276.
[15] RANZATO M A, POULTNEY C, CHOPRA S, et al. Efficient learning of sparse representations with an energy-based model[C]//Proceedings of the 19th International Conference on Neural Information Processing Systems. Kitakyushu, Japan:MIT Press, 2006:1137-1144.
[16] BENGIO Y. Learning deep architectures for AI[J]. Foundations and Trends in Machine Learning, 2009, 2(1):1-127.
[17] LECUN Y, BENGIO Y, HINTON G. Deep learning[J]. Nature, 2015, 521(7553):436-444.
[18] 周飞燕, 金林鹏, 董军. 卷积神经网络研究综述[J]. 计算机学报, 2017, 40(6):1229-1251. ZHOU Feiyan, JIN Linpeng, DONG Jun. Review of convolutional neural network[J]. Chinese Journal of Computers, 2017, 40(6):1229-1251.
[19] 王华斌, 韩旻, 王光辉, 等. 遥感影像要素提取的可变结构卷积神经网络方法[J]. 测绘学报, 2019, 48(5):583-596. DOI:10.11947/j.AGCS.2019.20180122. WANG Huabin, HAN Min, WANG Guanghui, et al. Surface features extraction in remote sensing images based on architecture-variant CNN[J]. Acta Geodaetica et Cartographica Sinica, 2019, 48(5):583-596. DOI:10.11947/j.AGCS.2019.20180122.
[20] FAN Dazhao, DONG Yang, ZHANG Yongsheng. Satellite image matching method based on deep convolutional neural network[J]. Journal of Geodesy and Geoinformation Science, 2019, 2(2):90-100. DOI:10.11947/j.JGGS.2019.0210.
[21] SILVER D, HUANG A, MADDISON C J, et al. Mastering the game of go with deep neural networks and tree search[J]. Nature, 2016, 529(7587):484-489.
[22] 余东行, 郭海涛, 张保明, 等. 级联卷积神经网络的遥感影像飞机目标检测[J]. 测绘学报, 2019, 48(8):1046-1058. DOI:10.11947/j.AGCS.2019.20180471. YU Donghang, GUO Haitao, ZHANG Baoming, et al. Aircraft detection in remote sensing images using cascade convolutional neural networks[J]. Acta Geodaetica et Cartographica Sinica, 2019, 48(8):1046-1058. DOI:10.11947/j.AGCS.2019.20180471.
[23] 唐秋华, 刘保华, 陈永奇, 等. 结合遗传算法的LVQ神经网络在声学底质分类中的应用[J]. 地球物理学报, 2007, 50(1):313-319. TANG Qiuhua, LIU Baohua, CHEN Yongqi, et al. Application of LVQ neural network combined with the genetic algorithm in acoustic seafloor classification[J]. Chinese Journal of Geophysics, 2007, 50(1):313-319.
[24] 纪雪. 基于多波束数据的海底底质及地形复杂度分类研究[D]. 青岛:国家海洋局第一海洋研究所, 2017. JI Xue. Classification of seabed sediment and terrain complexity based on multibeam data[D]. Qingdao:The First Institute of Oceanography, 2017.
[25] 付楠. 基于声呐图像特征的海底底质类型分类方法研究[D]. 哈尔滨:哈尔滨工程大学, 2019. FU Nan. Research on classification method of submarine substrate type based on characteristics of sonar image[D]. Harbin:Harbin Engineering University, 2019.
[26] HARALICK R M. Statistical and structural approaches to texture[J]. Proceedings of the IEEE, 1979, 67(5):786-804.
[27] SUN Wanxiao, KOLAPPAL A Z, GONG Peng. Two computation methods for detecting anisotropy in image texture[J]. Geographic Information Sciences, 2005, 11(2):87-96.
[28] 郑淑丹, 郑江华, 石明辉, 等. 基于分形和灰度共生矩阵纹理特征的种植型药用植物遥感分类[J]. 遥感学报, 2014, 18(4):868-886. ZHENG Shudan, ZHENG Jianghua, SHI Minghui, et al. Classification of cultivated Chinese medicinal plants based on fractal theory and gray level co-occurrence matrix textures[J]. Journal of Remote Sensing, 2014, 18(4):868-886.
[29] 郭德军, 宋蛰存. 基于灰度共生矩阵的纹理图像分类研究[J]. 林业机械与木工设备, 2005, 33(7):21-23. GUO Dejun, SONG Zhecun. A study on texture image classifying based on gray-level co-occurrence matrix[J]. Forestry Machinery & Woodworking Equipment, 2005, 33(7):21-23.
[30] DÍAZ J V M. Analysis of multibeam sonar data for the characterization of seafloor habitats[D]. New Brunswick:University of New Brunswick, 2000.
[31] ZHAO Jianhu, YAN Jun, ZHANG Hongmei, et al. Two self-adaptive methods of improving multibeam backscatter image quality by removing angular response effect[J]. Journal of Marine Science and Technology, 2017, 22(2):288-300.
[32] MARDIA K V. Measures of multivariate skewness and kurtosis with applications[J]. Biometrika, 1970, 57(3):519-530.
[33] GROENEVELD R A, MEEDEN G. Measuring skewness and kurtosis[J]. Journal of the Royal Statistical Society:Series D (The Statistician), 1984, 33(4):391-399.
[34] 刘学军, 龚健雅, 周启鸣, 等. 基于DEM坡度坡向算法精度的分析研究[J]. 测绘学报, 2004, 33(3):258-263. DOI:10.3321/j.issn:1001-1595.2004.03.014. LIU Xuejun, GONG Jianya, ZHOU Qiming, et al. A study of accuracy and algorithms for calculating slope and aspect based on grid digital elevation model (DEM)[J]. Acta Geodaetica et Cartographica Sinica, 2004, 33(3):258-263. DOI:10.3321/j.issn:1001-1595.2004.03.014.
[35] HINTON G E, SALAKHUTDINOV R R. Reducing the dimensionality of data with neural networks[J]. Science, 2006, 313(5786):504-507.
[36] KIRANYAZ S, INCE T, GABBOUJ M. Real-time patient-specific ECG classification by 1D convolutional neural networks[J]. IEEE Transactions on Biomedical Engineering, 2016, 63(3):664-675.
[37] CONGALTON R G. A review of assessing the accuracy of classifications of remotely sensed data[J]. Remote Sensing of Environment, 1991, 37(1):35-46.
[38] LEWIS K B. A reversal of throw and change of trend on the Wellington fault in Wellington Harbour[J]. New Zealand Journal of Geology and Geophysics, 1989, 32(2):293-298.