简介:[1]AiNanshan,1999.Makingforfractalphysiognomy.GeographyandTerritorialRes.,15(1):92-96.(inChinese)[2]AiNanshan,1993.FromMandelbrotlandscapetofractalphysiognomy.NatureJ.,16(1):13-17.(inChinese)[3]AiNanshan,ZhuZhijunetal.,1998.OnthestochasticnatureofexogenicprocessandthestabilityoffractionalBrownianlandscape.GeographicalResearch,17(1):23-29.(inChinese)[4]BBMandelbrot,1967.HowlongisthecoastofBritain?Statisticalself-similarityandfractaldimension.Science,150(3775):636-638.[5]ChenHui,GuoShichang,1997.ThemultifractalstudyofchangesofclimateinKunmingarea.ClimaticandEnvironmentResearch,2(4):261-268.(inChinese)[6]ChenYanguang,1999.Geography:thefailureofcomputionmotionandthegrowingupoffractalstudies.JournalofXinyangNormalUniversity(NaturalScienceEdition),12(3):310-314.(inChinese)[7]ChenYanguang,1997.Onfractalsandtouristlandscape.HumanGeography,12(1):62-66.(inChinese)[8]ChenYanguang,WangYimin,1999.Fractal,1/ffluctuationandtheaestheticessenceoftouristresorts.ExplorationofNature,18(3):51-54.(inChinese)[9]ChenYanguang,LiBaolin,2003.StudiesofthefractalnetworkcompositionofriversinJilinprovince,China.AdvanceinEarthSciences,18(2):178-184.(inChinese)[10]DingYizhong,LouYong,1998.Applicationoffractaltheoryintheevaluationontransportationnetwork.JournalofShanghaiMaritimeUniversity,19(4):7-12.(inChinese)[11]DingWenfeng,DingDengshan,2002.ThefractalfeaturesofsoilgranulestructurebeforeandaftervegetationdestructiononLoessPlateau.GeographicalResearch,21(6):700-706.(inChinese)[12]FengJinliang,ZhengLi,1997.Thesimpleanalysisofgeographicsignificanceoffractaldimensionofcoastline.MarineGeologyandQuaternaryGeology,17(1):35-51.(inChinese)[13]HuangGuilan,ZhengZhaobao,1995.Theapplicationoffuzzyclusteranalysisinimagetexturebasedonfractal.JournalofWuha
简介:针对亚热带地区树种丰富,树种间相似度大对树种识别带来的问题,本研究以福建省三明市莘口镇格氏栲自然保护区附近13种常见树种的冠层实测高光谱数据为例,采用一种分层分析方法,探讨不同光谱类别的树种识别精度以及树种识别的最佳波段。首先,对原始光谱进行变换处理,包括一阶微分、二阶微分、对数一阶微分、包络线去除和植被指数;其次,通过分析选择出13种树种各光谱类别的差异显著波段;最后,利用逐步判别法对选择的差异显著波段进一步降维,判断不同光谱类别的树种识别精度并找出识别13种树种的最佳波段。结果表明:光谱变换能有效地提高树种的识别精度,尤其是对数一阶微分光谱,总识别精度高达98.7%;对于原始光谱,近红外波段(760~1300nm)的树种识别能力更强,对于变换光谱,可见光波段(350~760nm)的树种识别效果更佳;不同光谱类别之间具体的显著性差异波段存在很大差别,原始光谱与变换光谱之间仅在绿光波段(500~600nm)有少量相同的显著性差异波段,此研究成果可为亚热带地区树种识别提供参考。