[1]杜伟东,李海森,魏玉阔,等.基于SVM的决策融合鱼类识别方法[J].哈尔滨工程大学学报,2015,(05):623-627.[doi:10.3969/j.issn.1006-7043.201403083]
 DU Weidong,LI Haisen,WEI Yukuo,et al.Decision fusion fish identification using SVM and its experimental study[J].hebgcdxxb,2015,(05):623-627.[doi:10.3969/j.issn.1006-7043.201403083]
点击复制

基于SVM的决策融合鱼类识别方法(/HTML)
分享到:

《哈尔滨工程大学学报》[ISSN:1006-6977/CN:61-1281/TN]

卷:
期数:
2015年05期
页码:
623-627
栏目:
出版日期:
2015-05-25

文章信息/Info

Title:
Decision fusion fish identification using SVM and its experimental study
作者:
杜伟东12 李海森12 魏玉阔12 徐超12
1. 哈尔滨工程大学 水声技术重点实验室, 黑龙江 哈尔滨 150001;
2. 哈尔滨工程大学 水声工程学院, 黑龙江 哈尔滨 150001
Author(s):
DU Weidong12 LI Haisen12 WEI Yukuo12 XU Chao12
1. Acoustic Science and Technology Laboratory, Harbin Engineering University, Harbin 150001, China;
2. College of Underwater Acoustic Engineering, Harbin Engineering University, Harbin 150001, China
关键词:
鱼类识别多方位决策融合支持向量机小波包变换离散余弦变换
分类号:
S932.4;P745
DOI:
10.3969/j.issn.1006-7043.201403083
文献标志码:
A
摘要:
为解决基于声学散射数据的高精度鱼类识别问题, 提出一种基于SVM的多方位声散射数据决策层融合的鱼类识别方法。利用小波包变换(WPT)和离散余弦变换(DCT)方法对多方位声散射数据进行特征提取, 并进行特征降维处理。然后采用SVM分类器对每个方位提取的特征做出多次决策, 并输出最终识别结果。采用3种不同鱼类作为研究对象, 设计了可靠的获取多方位声散射数据的实验方案, 给出不同方位数量条件下, 基于WPT和DCT特征量的识别率。理论分析及实验数据处理结果表明, 随着方位数量的增加, 总体识别率呈升高的趋势, 基于SVM的多方位声散射数据决策层融合方法可以有效提高识别率至90%以上。

参考文献/References:

[1] ROBOTHAM H, BOSCH P, CARLOS J, et al. Acoustic identification of small pelagic fish species in Chile using support vector machines and neural networks[J]. Fisheries Research, 2010, 102(1/2): 115-122.
[2] LUNDGREN B, NIELSEN J R. A method for the possible species discrimination of juvenile gadoids by broad-bandwidth backscattering spectra vs. angle of incidence[J]. ICES Journal of Marine Science, 2008, 65(4): 581-593.
[3] SLAS-CITAL A, ATKINS P R, FOO K Y. Broadband amplitude and phase sonar calibration using LFM pulses for high-resolution study of hard and soft acoustic targets[C]// OCEANS 2011, Spain, 2011: 1-8.
[4] LIU G, JAYATHILAKE P G, KHOO B C. Conformal mapping for the Helmholtz equation Acoustic wave scattering by a two dimensional inclusion with irregular shape in an ideal fluid[J]. Journal of the Acoustical Society of America, 2012, 131(2): 1055-1065.
[5] CARTMILL J, WACHOWSKI N, AZIMI-SADJADI M R. Buried underwater object classification using a collaborative multi-aspect classifier[J]. IEEE Journal of Oceanic Engineering, 2009, 34(1): 32-44.
[6] 杜伟东, 李海森, 陈宝伟, 等. 一种基于声散射特性的有鳔鱼特征获取方法[J]. 应用声学, 2014, 33(6): 505-511.DU Weidong, LI Haisen, CHEN Baowei, et al. Features acquisition of fish with swim bladder baesd on acoustic scattering characteristics[J]. Journal of Applied Acoustics, 2014, 33(6): 505-511.
[7] 张汝波, 张鑫瑜, 李雪耀. 利用支持向量机和高阶累量实现飞机类型识别[J]. 哈尔滨工程大学学报, 2010, 31(3): 366-370.ZHANG Rubo, ZHANG Xinyu, LI Xueyao. Recognizing aircraft type using a support vector machine and a higher order cumulant[J]. Journal of Harbin Engineering University, 2010,31(3): 366-370.
[8] AZIMI-SADJADI M R, YAO D, JAMSHIDI A, et al. Underwater target classification in changing environments using an adaptive feature mapping[J]. IEEE Transactions on Neural Networks, 2002, 13(5): 1099-1111.
[9] LI D, AZIMI-SADJADI M R, ROBINSON M. Comparison of different classification algorithms for underwater target discrimination[J]. IEEE Transactions on Neural Networks, 2004, 15(1): 189-194.
[10] 石强, 陈凤娥, 梅天灿, 等. SVM后验概率结合改进多尺度MRF的遥感影像分割方法[J]. 武汉大学学报信息科学版, 2013, 38(2): 195-199.SHI Qiang, CHEN Feng’e, MEI Tiancan, et al. Remote sensing image segmentation based on SVM posterior probability and improved multi-scale MRF [J]. Geomatics and Information Science of Wuhan University, 2013, 38(2): 195-199.

备注/Memo

备注/Memo:
收稿日期:2014-3-26;改回日期:。
基金项目:国家自然科学基金资助项目(41306038);水声技术重点实验室基金资助项目(9140C200105120C2001).
作者简介:杜伟东(1984-),男,博士研究生;李海森(1962-),男,教授,博士生导师.
通讯作者:李海森,E-mail:hsenli@126.com.
更新日期/Last Update: 2015-06-15