联合支持向量机和增强学习算法的多波束声学底质分类

doi:10.11947/j.AGCS.2021.20200556

测绘学报 ›› 2021, Vol. 50 ›› Issue (7): 972-981.doi: 10.11947/j.AGCS.2021.20200556

联合支持向量机和增强学习算法的多波束声学底质分类

纪雪¹, 唐秋华^2,3, 陈义兰², 李杰², 丁德秋³

1. 测绘遥感信息工程国家重点实验室, 湖北武汉 430079;
2. 自然资源部第一海洋研究所, 山东青岛 266061;
3. 山东科技大学测绘科学与工程学院, 山东青岛 266590

收稿日期:2020-11-18 修回日期:2021-03-04 发布日期:2021-08-13
通讯作者: 唐秋华 E-mail:tangqiuhua@fio.org.cn
作者简介:纪雪(1989-),女,博士生,研究方向为海洋测绘。E-mail:jixuesdqd@whu.edu.cn
基金资助:
国家自然科学基金（41876111）；国家海洋局极地考察办公室项目（IRASCC2020-2022）

Multibeam acoustic seabed classification combining SVM and adaptive boosting algorithm

JI Xue¹, TANG Qiuhua^2,3, CHEN Yilan², LI Jie², DING Deqiu³

1. State Key Laboratory of Information Engineering in Surveying, Mapping and Remote Sensing, Wuhan 430079, China;
2. First Institute of Oceanography, Ministry of Natural Resources, Qingdao 266061, China;
3. College of Geodesy and Geomatics, Shandong University of Science and Technology, Qingdao 266590, China

Received:2020-11-18 Revised:2021-03-04 Published:2021-08-13
Supported by:
The National Natural Science Foundation of China (No. 41876111);The Chinese Arctic and Antarctic Administration Project (No. IRASCC2020-2022)

摘要/Abstract

摘要： 基于多波束的声学底质分类是近年来快速发展起来的新型海底底质探测技术。针对多波束声学底质分类中底质类型多样化、类型之间差异较小等多分类难点问题，本文提出一种GA-SVM-AdaBoost算法。利用自适应性和全局搜索能力强的遗传算法（genetic algorithm，GA）去优化支持向量机（support vector machines，SVM），以获得最优模型初始参数，并将多个GA优化后的SVM作为弱分类器组成AdaBoost强分类器。对胶州湾采集到的多波束反向散射强度数据，经过精细处理后生成海底声呐镶嵌图；构建SVM-RFE-CBR算法从提取的36维声强空间特征中筛选出10维优势特征，将其输入到GA-SVM-AdaBoost模型中进行分类识别。通过与SVM、GA-SVM、基于单层决策树的AdaBoost分类模型对比，GA-SVM-AdaBoost算法的总体分类精度高达92.19%，优于另外3种模型，证明GA-SVM-AdaBoost分类模型可有效应用于高精度海底底质类型识别。

关键词: 多波束测深系统, 声学底质分类, 支持向量机, 遗传算法, AdaBoost算法

Abstract: As a new technology, multibeam acoustic classification has been rapidly developed in recent years. A seabed sediment classification approach, GA-SVM-AdaBoost algorithm, is proposed by using the genetic algorithm (GA) optimized support vector machines (SVM) classifier as the AdaBoost weak classifier to solve the multi-classification problem in multibeam acoustic seabed classification. The sonar mosaic is obtained from multibeam echo sounder backscatter data collected in the Jiaozhou Bay within fine processing. The 10 dimensions advantage features are selected by SVM-RFE-CBR algorithm before input GA-SVM-AdaBoost classification model. Compared with SVM, GA-SVM and AdaBoost based on single-layer decision tree, the classification results of GA-SVM-AdaBoost algorithm are more satisfactory. The total classification accuracy is as high as 92.19%, which is better than the other three models. It is proved that the proposed method can be effectively applied to high precision seabed sediment identification.

Key words: multibeam sounding system, acoustic seabed classification, support vector machine, genetic algorithm, AdaBoost algorithm

中图分类号:

P229

纪雪, 唐秋华, 陈义兰, 李杰, 丁德秋. 联合支持向量机和增强学习算法的多波束声学底质分类[J]. 测绘学报, 2021, 50(7): 972-981.

JI Xue, TANG Qiuhua, CHEN Yilan, LI Jie, DING Deqiu. Multibeam acoustic seabed classification combining SVM and adaptive boosting algorithm[J]. Acta Geodaetica et Cartographica Sinica, 2021, 50(7): 972-981.

参考文献

[1] ANDERSON J T, VAN HOLLIDAY D, KLOSER R, et al. Acoustic seabed classification: current practice and future directions[J]. ICES Journal of Marine Science, 2008, 65(6): 1004-1011.
[2] 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.
[3] BROWN C J, MITCHELL A, LIMPENNY D S, et al. Mapping seabed habitats in the Firth of Lorn off the west coast of Scotland: evaluation and comparison of habitat maps produced using the acoustic ground-discrimination system, RoxAnn, and sidescan sonar[J]. ICES Journal of Marine Science, 2005, 62(4): 790-802.
[4] LARK R M, DOVE D, GREEN S L, et al. Spatial prediction of seabed sediment texture classes by cokriging from a legacy database of point observations[J]. Sedimentary Geology, 2012, 281: 35-49.
[5] 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.
[6] WILKEN D, FELDENS P, WUNDERLICH T, et al. Application of 2D Fourier filtering for elimination of stripe noise in side-scan sonar mosaics[J]. Geo-Marine Letters, 2012, 32(4): 337-347.
[7] FAJARYANTI R, KANG M. A preliminary study on seabed classification using a scientific echosounder[J]. Bulletin of the Korean Society of Fisheries Technology, 2019, 55(1): 39-49.
[8] OJEDAG Y, GAYES P T, VAN DOLAH R F, et al. Spatially quantitative seafloor habitat mapping: example from the northern South Carolina inner continental shelf[J]. Estuarine, Coastal and Shelf Science, 2004, 59(3): 399-416.
[9] 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.
[10] 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.
[11] ZHOU Xinghua, CHEN Yongqi. Seafloor sediment classification based on multibeam sonar data[J]. Geo-Spatial Information Science, 2004, 7(4): 290-296.
[12] LI Jin, TRAN M, SIWABESSY J. Selecting optimal random forest predictive models: a case study on predicting the spatial distribution of seabed hardness[J]. PLoS ONE, 2016, 11(2):1490-1508.
[13] HERKVL, 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(1): 57-71.
[14] TURNER J A, BABCOCK R C, HOVEY R, et al. Can single classifiers be as useful as model ensembles to produce benthic seabed substratum maps[J]. Estuarine, Coastal and Shelf Science, 2018, 204(2): 149-163.
[15] JI Xue, YANG Bisheng, TANG Qiuhua. Seabed sediment classification using multibeam backscatter data based on the selecting optimal random forest model[J]. Applied Acoustics, 2020, 167: 107387.
[16] 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, State Oceanic Administra-tion, 2017.
[17] BERTHOLD T, LEICHTER A, ROSENHAHN B, et al. Seabed sediment classification of side-scan sonar data using convolutional neural networks[C]//Proceedings of 2017 IEEE Symposium Series on Computational Intelligence. Honolulu, HI, USA: IEEE, 2017:1-8.
[18] 阳凡林,朱正任, 李家彪, 等. 利用深层卷积神经网络实现地形辅助的多波束海底底质分类[J]. 测绘学报, 2021, 50(1): 71-84. DOI:10.11947/j.AGCS.2021.20200065. YANG Fanlin, ZHU Zhengren, LI Jiabiao, et al. 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. DOI:10.11947/j.AGCS.2021.20200065.
[19] SCHAPIRE R E. A brief introduction to boosting[C]//Proceedings of the 16th International Joint Conference on Artificial Intelligence. Stockholm, Sweden: Morgan Kaufmann Publishers Inc., 1999: 1401-1406.
[20] FREUND Y, SCHAPIRE R E. A decision-theoretic generalization of on-line learning and an application to boosting[J]. Journal of Computer and System Sciences, 1997, 55(1): 119-139.
[21] VAPNIK V N. The nature of statistical learning theory[M]. New York: Springer, 1995.
[22] VAPNIK V N. An overview of statistical learning theory[J]. IEEE Transactions on Neural Networks, 1999, 10(5): 988-999.
[23] BURGES C J C. A tutorial on support vector machines for pattern recognition[J]. Data Mining and Knowledge Discovery, 1998, 2(2): 121-167.
[24] SMOLA A J, SCHÖLKOPF B. A tutorial on support vector regression[J]. Statistics and Computing, 2004, 14(3): 199-222.
[25] HOLLAND J H. Genetic algorithms[J]. Scientific American, 1992, 267(1): 66-73.
[26] MICHALOPOULOU Z H, ALEXANDROU D, DE MOU-STIER C. Application of neural and statistical classifiers to the problem of seafloor characterization[J]. IEEE Journal of Oceanic Engineering, 1995, 20(3): 190-197.
[27] HELLEQUIN L, BOUCHER J M, LURTON X. Processing of high-frequency multibeam echo sounder data for seafloor characterization[J]. IEEE Journal of Oceanic Engineering, 2003, 28(1): 78-89.
[28] FONSECA L, BROWN C, CALDER B, et al. Angular range analysis of acoustic themes from Stanton Banks Ireland: a link between visual interpretation and multibeamechosounder angular signatures[J]. Applied Acoustics, 2009, 70(10): 1298-1304.
[29] 严俊, 张红梅, 赵建虎, 等. 多波束声呐后向散射数据角度响应模型的改进算法[J]. 测绘学报, 2016, 45(11): 1301-1307. DOI: 10.11947/j.AGCS.2016.20160169. YAN Jun, ZHANG Hongmei, ZHAO Jianhu, et al. Study on improvement of multibeam backscatter angular response model[J]. Acta Geodaetica et Cartographica Sinica, 2016, 45(11): 1301-1307. DOI: 10.11947/j.AGCS.2016.20160169.
[30] LECOURS V, DOLAN M F J, MICALLEF A, et al. A review of marine geomorphometry, the quantitative study of the seafloor[J]. Hydrology and Earth System Sciences, 2016, 20(8): 3207-3244.
[31] 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. 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. DOI: 10.11947/j.AGCS.2016.20160466.
[32] 胡韦伟, 汪荣贵, 方帅, 等. 基于双边滤波的Retinex图像增强算法[J]. 工程图学学报, 2010, 31(2): 104-109. HU Weiwei, WANG Ronggui, FANG Shuai, et al. Retinex algorithm for image enhancement based on bilateral filtering[J]. Journal of Engineering Graphics, 2010, 31(2): 104-109.
[33] RAHMAN Z, JOBSON D J, WOODELL G A. Retinex processing for automatic image enhancement[J]. Journal of Electronic Imaging, 2004, 13(1): 100-110.
[34] YAN Ke, ZHANG D. Feature selection and analysis on correlated gas sensor data with recursive feature elimination[J]. Sensors and Actuators B: Chemical, 2015, 212: 353-363.
[35] SCHAPIRE R E, SINGER Y. BoosTexter: A boosting-based system for text categorization[J]. Machine Learning, 2000, 39(2-3): 135-168.
[36] ZHU Ji, ZOU Hui, ROSSET S, et al. Multi-class AdaBoost[J]. Statistics and its Interface, 2009, 2(3): 349-360.

联合支持向量机和增强学习算法的多波束声学底质分类

Multibeam acoustic seabed classification combining SVM and adaptive boosting algorithm

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]	王丹菂, 邢帅, 徐青, 林雨准, 李鹏程. 单频机载激光测深海陆回波自动分类方法[J]. 测绘学报, 2022, 51(5): 750-761.
[2]	刘远刚, 李少华, 蔡永香, 何贞铭, 马潇雅, 李鹏程, 郭庆胜, 何宗宜. 移位安全区约束下的建筑物群移位免疫遗传算法[J]. 测绘学报, 2021, 50(6): 812-822.
[3]	周建伟, 吴一全. MRELBP特征、Franklin矩和SVM相结合的遥感图像建筑物识别方法[J]. 测绘学报, 2020, 49(3): 355-364.
[4]	余岸竹, 刘冰, 邢志鹏, 杨帆, 杨其淼. 面向高光谱影像分类的显著性特征提取方法[J]. 测绘学报, 2019, 48(8): 985-995.
[5]	王华斌, 韩旻, 王光辉, 李玉. 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.
[6]	闫利, 谭骏祥, 刘华, 陈长军. 融合遗传算法和ICP的地面与车载激光点云配准[J]. 测绘学报, 2018, 47(4): 528-536.
[7]	张海波, 汪长城, 朱建军, 付海强. 利用ESAR极化数据的复杂地形区森林地上生物量估算[J]. 测绘学报, 2018, 47(10): 1353-1362.
[8]	张立华, 刘现鹏, 贾帅东, 史岩. 一种线面组合的水下地形匹配算法[J]. 测绘学报, 2018, 47(10): 1406-1414.
[9]	马超, 孙群, 陈换新, 徐青, 杨辉. 志愿者地理信息中天桥的自动识别方法[J]. 测绘学报, 2017, 46(2): 246-252.
[10]	许夙晖, 慕晓冬, 赵鹏, 马骥. 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.
[11]	付仲良, 杨元维, 高贤君, 赵星源, 范亮. 道路网多特征匹配优化算法[J]. 测绘学报, 2016, 45(5): 608-615.
[12]	周鑫鑫, 吴长彬, 孙在宏, 丁远, 贺涛. 小规模地理场景中点要素三维注记优化配置算法[J]. 测绘学报, 2016, 45(12): 1476-1484.
[13]	李慧, 张金区, 曹阳, 王兴芳. 端元可变非线性混合像元分解模型[J]. 测绘学报, 2016, 45(1): 80-86.
[14]	张春森, 郑艺惟, 黄小兵, 崔卫红. 高光谱影像光谱-空间多特征加权概率融合分类[J]. 测绘学报, 2015, 44(8): 909-918.
[15]	刘丽峰, 杨飞, 张树清, 孔维华, 王殷行. 应用人工免疫算法快速规划3D TF/TA²航线[J]. 测绘学报, 2015, 44(4): 462-470.