全極化合成孔徑雷達(dá)多視圖像的極化特征分析
Statistical analysis of multilook signatures for polarimetric SAR image
-
摘要: 該文基于乘性相干斑模型,對(duì)全極化合成孔徑雷達(dá)多視圖像的極化特征參數(shù)(同極化比、交叉極化比和同極化相位差)進(jìn)行了分析,利用實(shí)際的模擬數(shù)據(jù)對(duì)特征參數(shù)的PDF進(jìn)行了擬合測(cè)試,分析了多視極化特征參數(shù)的統(tǒng)計(jì)特性對(duì)極化相干斑抑制算法的影響。利用極化比和相位差的特點(diǎn),對(duì)地物進(jìn)行分類,所得分類精度與利用最優(yōu)極化分類法的分類精度具有可比性。
-
關(guān)鍵詞:
- 極化合成孔徑雷達(dá); 多視; 相干斑; 極化特征; 分類
Abstract: Based on the multiplication speckle model, the statistical characteristics of the polarimetric signatures are analyzed,which include co-polar and cross-polar ratios and co-polar phase difference for multi-look polarimetric SAR data. With the simulated data, the effect of the estimated accuracy of the polarimetric signatures on the performance of the algorithm for suppressing the speckle is analyzed. in addition, based on the features of the polarimetric signatures, applying them on the Bayes classifier, the accuracy of classification can be compared with the optimum polarimetric classifier. -
L.M. Novak, M. C. Burl, Optimal speckle reduction in polarimetric SAR imagery, IEEE Trans.on Aerospace and Electronic Systems, 1990, AES-26(2), 293-304.[2]Jong-Sen Lee, M. R. Grunes, Speckle reduction in multipolarization, multifrcquency SAR imagery,IEEE Trans. on Geoscicence and Remote Sensing, 1991, GRS-29(4), 535-544.[3]I.R. Joughin, D. P. Winebrenner, Probability density functions for multilook polarimetric signatrures, IEEE Trans. on Geoscience and Remote Sensing, 1994, GRS-32(3), 562-574.[4]M.C. Teich, P. Diament, Multiply stochastic representations for K distributions and their Poisson transforms, J. Opt. Soc. Amer., 1989, 6(1), 80-91.[5]L. M. Novak, Generating correlated Weibull random variables for digital simulations. In Proc. of IEEE Decision and Control Conference, San Diego, CA, 1973, 156-160.[6]L.M. Novak, R. T. Shin, Identification of terrain cover using the optimum polarimetric classifier,Journal of Electromagnetic Waves and Application, 1987, 2(2), 171-194. -
計(jì)量
- 文章訪問(wèn)數(shù): 2512
- HTML全文瀏覽量: 76
- PDF下載量: 704
- 被引次數(shù): 0
下載:
