首页 | 本学科首页   官方微博 | 高级检索  
     检索      

基于小波变换的油膜光谱特征分析
引用本文:李颖,刘丙新,李宝玉,陈铎.基于小波变换的油膜光谱特征分析[J].光谱学与光谱分析,2012,32(7):1923-1927.
作者姓名:李颖  刘丙新  李宝玉  陈铎
作者单位:大连海事大学环境信息研究所,辽宁 大连 116026
基金项目:国家自然科学基金项目(41071260,41171329);国家海洋局公益项目(2012418012-02);辽宁省教育厅项目(2009T010)资助
摘    要:水上溢油光谱作为高光谱遥感目标识别与分类的参考依据,在溢油识别与厚度区分等方面具有重要的研究意义。通过测量厚度范围为1.0~127 μm间的轻柴油的20组光谱曲线,计算其反射率光谱曲线随厚度变化的特征,并利用db4小波对反射率数据进行处理,突出光谱奇异性及其位置与奇异值随油膜厚度变化的特征。在研究范围内的油膜光谱反射率高于水体,但反射率值与油膜厚度间无固定增减关系。在其厚度小于6 μm时反射率光谱曲线无明显特征,厚度大于6 μm后在388 nm附近存在区别于水体的反射峰特征且随厚度增加特征愈加突出。油膜光谱小波分析后的细节系数在388~393 nm内表现出明显的奇异性,并且奇异位置随厚度增加向短波方向移动、奇异极值增大。研究证实了小波分析在确定油膜光谱特征位置与变化研究中的积极作用,并发现了紫外-可见光短波范围内的光谱特征,为紫外遥感进行溢油识别提供了科学依据。

关 键 词:小波分析  溢油  光谱  遥感  
收稿时间:2011-12-12

Analysis of Spectral Characteristics of Oil Film on Water Based on Wavelet Transform
LI Ying,LIU Bing-xin,LI Bao-yu,CHEN Duo.Analysis of Spectral Characteristics of Oil Film on Water Based on Wavelet Transform[J].Spectroscopy and Spectral Analysis,2012,32(7):1923-1927.
Authors:LI Ying  LIU Bing-xin  LI Bao-yu  CHEN Duo
Institution:Environmental Information Institute, Dalian Maritime University, Dalian 116026, China
Abstract:The diagnostic features are the basis to detect and characterize the oil film on water through optical remote sensing. This work shows the results of lab spectral measurements of light diesel oil with thickness ranged 1.0~127 μm. A wavelet transform were performed to the reflectance, and the singularity (388~393 nm) was explored as the indicators of oil film thickness. The results indicate that the reflectance of light diesel oil film is higher than that of water in the range from 350~2 500 nm. There is a reflectance peak near 388 nm when the thickness of oil film is larger than 6 μm, however, no distinguished features could be recognized when oil films were thinner than 6 μm. The wavelet coefficients of the fifth decomposition level by applying Daubechies 4 (db4) mother wavelets proved successful for identifying the singularity of oil film’s reflectance spectra and its accurate position. With the thickness lager than 6 μm, the detail coefficients performed an abrupt change within the range of 388~393 nm, and became more violent while oil films’ thickness increased. This research demonstrated that oil films on water with different thickness could be distinguished based on wavelet detail coefficients, with important implications for detection of oils on water using UV and short wave optical remote sensing.
Keywords:Wavelet analysis  Oil spill  Spectrum  Remote sensing  
本文献已被 CNKI 等数据库收录!
点击此处可从《光谱学与光谱分析》浏览原始摘要信息
点击此处可从《光谱学与光谱分析》下载免费的PDF全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号