国土资源遥感
國土資源遙感
국토자원요감
REMOTE SENSING FOR LAND & RESOURCES
2014年
2期
11-18
,共8页
严婷婷%边红枫%廖桂项%盛连喜%张吉顺%高明辉
嚴婷婷%邊紅楓%廖桂項%盛連喜%張吉順%高明輝
엄정정%변홍풍%료계항%성련희%장길순%고명휘
森林湿地%遥感技术%水文反演%湿地分类%湿地识别
森林濕地%遙感技術%水文反縯%濕地分類%濕地識彆
삼림습지%요감기술%수문반연%습지분류%습지식별
forested wetland%remote sensing techniques%hydrological inversion%wetlands classification%wetlands mapping
森林湿地是湿地的重要组成部分,因其群落结构复杂,从景观尺度上对其进行识别是湿地研究的难点之一。在分析国内外相关科学文献的基础上,从地理生态环境、影像特征及影像信息处理单元角度分析了森林湿地遥感信息提取方法的研究现状,总结了当前森林湿地遥感提取的特点,并在此基础上初步预测未来森林湿地遥感信息提取方法的发展趋势。结果表明:以水文特征反演为依据,认为基于地理生态环境的提取方法包括基于水文地貌学方法、基于光学遥感和微波遥感的分类方法;基于影像特征提取方法包括基于雷达散射特征决策树、随机森林决策树及航片目视解译法;基于影像信息处理单元角度方法主要有基于像元和对象分类2种方法。
森林濕地是濕地的重要組成部分,因其群落結構複雜,從景觀呎度上對其進行識彆是濕地研究的難點之一。在分析國內外相關科學文獻的基礎上,從地理生態環境、影像特徵及影像信息處理單元角度分析瞭森林濕地遙感信息提取方法的研究現狀,總結瞭噹前森林濕地遙感提取的特點,併在此基礎上初步預測未來森林濕地遙感信息提取方法的髮展趨勢。結果錶明:以水文特徵反縯為依據,認為基于地理生態環境的提取方法包括基于水文地貌學方法、基于光學遙感和微波遙感的分類方法;基于影像特徵提取方法包括基于雷達散射特徵決策樹、隨機森林決策樹及航片目視解譯法;基于影像信息處理單元角度方法主要有基于像元和對象分類2種方法。
삼림습지시습지적중요조성부분,인기군락결구복잡,종경관척도상대기진행식별시습지연구적난점지일。재분석국내외상관과학문헌적기출상,종지리생태배경、영상특정급영상신식처리단원각도분석료삼림습지요감신식제취방법적연구현상,총결료당전삼림습지요감제취적특점,병재차기출상초보예측미래삼림습지요감신식제취방법적발전추세。결과표명:이수문특정반연위의거,인위기우지리생태배경적제취방법포괄기우수문지모학방법、기우광학요감화미파요감적분류방법;기우영상특정제취방법포괄기우뢰체산사특정결책수、수궤삼림결책수급항편목시해역법;기우영상신식처리단원각도방법주요유기우상원화대상분류2충방법。
Forested wetland is an important type of wetlands. Because of its complex community structure, mapping forested wetland becomes one of challenging issues in wetland research. In this paper, research conclusions published in relevant literatures were analyzed, and then a review of methods for mapping forested wetlands was made from the angles of geographical biological environment, images characteristics and images mapping units. In addition, the basic features of different mapping methods of forested wetlands by remote sensing were presented, and some possible trends of future research were preliminarily predicted. According to understory hydrological inversion, the mapping methods based on geographical biological environment mainly include the hydrogeomorphy-based method and the optical and microwave remote sensing -based classification method, whereas methods based on features of images include decision tree based on radar statistical backscatter characteristics, random forest decision tree and aerial photograph interpretation, which provide data and technical supports for tapping comparing and assessing information, thus conducive to further development of mapping methods. From the angle of imagery interpretation units, the forested wetland mapping methods have undergone the development from per - pixel classification to object-oriented classification.