人民黄河
人民黃河
인민황하
Yellow River
2014年
7期
45-47,94
,共4页
和海霞%汤童%田卫方
和海霞%湯童%田衛方
화해하%탕동%전위방
冰凌%环境减灾卫星%决策树分类%黄河宁夏段
冰凌%環境減災衛星%決策樹分類%黃河寧夏段
빙릉%배경감재위성%결책수분류%황하저하단
ice slush%HJ-1 data%decision tree classification%Ningxia section of Yellow River
凌汛给沿河居民生命财产安全带来隐患,利用。感技术开展凌情监测对防灾减灾工作具有重要的现实意义。针对冰凌。感监测业务中存在的易受多种地物影响、自动化程度低、精度低等难题,基于 HJ -1 CCD 数据,在主要地物光谱特征和影像特征分析的基础上,采用决策树分类算法自动识别冰凌,总体精度为92.38%,kappa 系数为0.7249。2012年、2013年黄河冰凌发展关键期的监测实践表明,HJ -1 CCD 数据为冰凌监测提供了一种稳定有效的数据源。
凌汛給沿河居民生命財產安全帶來隱患,利用。感技術開展凌情鑑測對防災減災工作具有重要的現實意義。針對冰凌。感鑑測業務中存在的易受多種地物影響、自動化程度低、精度低等難題,基于 HJ -1 CCD 數據,在主要地物光譜特徵和影像特徵分析的基礎上,採用決策樹分類算法自動識彆冰凌,總體精度為92.38%,kappa 繫數為0.7249。2012年、2013年黃河冰凌髮展關鍵期的鑑測實踐錶明,HJ -1 CCD 數據為冰凌鑑測提供瞭一種穩定有效的數據源。
릉신급연하거민생명재산안전대래은환,이용。감기술개전릉정감측대방재감재공작구유중요적현실의의。침대빙릉。감감측업무중존재적역수다충지물영향、자동화정도저、정도저등난제,기우 HJ -1 CCD 수거,재주요지물광보특정화영상특정분석적기출상,채용결책수분류산법자동식별빙릉,총체정도위92.38%,kappa 계수위0.7249。2012년、2013년황하빙릉발전관건기적감측실천표명,HJ -1 CCD 수거위빙릉감측제공료일충은정유효적수거원。
Ice flood resulting from ice slush threatens the health and property safety of human along the Yellow River. Ice slush monitoring using re-mote sensing data is useful for disaster prevention and disaster reduction. Decision tree classification method was used to identify ice slush using HJ-1 based on the spectral and spatial distribution characteristic of main surface features. Its aim was to solve the low classification precision caused by various surface feature and low degree of automation. Its overall decision was 92. 38 percent and Kappa coefficient was 0. 7249. The application re-sults of ice slush monitoring in key developing stage from winter of 2012 to spring of 2013 indicated that HJ-1 CCD is a useful and stable data.