排序方式: 共有61条查询结果,搜索用时 31 毫秒
31.
支持向量机回归SVR(Support Vector Regression)方法作为叶面积指数反演的一种新思路,在LAI反演中具有一定的应用价值和前景,但SVR算法中惩罚系数C、核函数宽度参数g、不敏感损失函数参数ε的取值对回归精度有显著的影响。本文提出了一种基于人工蜂群算法ABC(Artificial Bee Colony)优化SVR参数的遥感影像叶面积指数反演方法。研究数据为美国土壤水分实验(SMEX02)2002年LAI实测数据和同期的Landsat 7 ETM+地表反射率数据,为了验证ABC算法优化SVR各个参数对反演精度的影响,建立了未优化参数(SVR)、优化单个参数(ABC-SVR-C,ABC-SVR-g,ABC-SVR-ε)、优化3个参数(ABC-SVR)的3类LAI反演模型,并比较了其回归拟合精度。在此基础上,分析了3个关键参数对LAI反演模型精度的敏感性,并对ABC算法优化SVR模型的精度进行显著性检验。研究表明:(1)相比未优化参数模型,ABC算法优化模型具有更高的反演精度,优化3个参数优于优化单个参数,回归直线斜率k达到0.797、决定系数r2达到0.775。(2)SVR的3个关键参数对模型精度都有影响,相较参数C和g,参数ε引起模型精度的不确定性更高。(3)95%的置信区间下,ABC-SVR模型与SVR模型的回归直线斜率k、r2、RMSE的差异显著性检验P值均小于0.005,ABC算法显著改善了SVR模型的精度。 相似文献
32.
33.
34.
35.
青藏高原小嵩草高寒草甸返青期遥感识别方法筛选 总被引:3,自引:1,他引:2
小嵩草高寒草甸是青藏高原的主要植被类型,研究其返青期识别方法对于模拟及预测青藏高原植被物候变化具有重要意义。常用的植被返青期遥感识别方法主要是先对遥感植被指数原始时序数据进行拟合去噪声再求取返青期,各种方法对研究区域、研究经验、参数设置、函数初值设置等有很强的依赖性。为避免返青期识别方法在曲线拟合时对参数初值的依赖性和陷入局部最优解,本文引入了模拟退火算法对双高斯和双逻辑斯蒂函数进行参数优化,并分别对基于以上两种函数及多项式拟合的植被指数时序曲线进行对比,从而选出最佳拟合方法,最后采用最大斜率阈值法、动态阈值法和曲率法识别返青期。利用青藏高原小嵩草高寒草甸34个样本点的返青期地面观测数据及相应的8 km分辨率的NOAA归一化差值植被指数(NDVI)时序数据对以上各种组合的返青期遥感识别方案进行了测试,并选取了153个遥感实验点求取了近30年(1982年—2011年)青藏高原小嵩草高寒草甸的返青期,结果表明:采用双高斯函数拟合的NDVI曲线与原始NDVI时序数据最为接近,在此基础上采用最大斜率阈值法识别的小嵩草高寒草甸返青期及其变化趋势与地面物候观测结果最为一致;同时发现近30年青藏高原小嵩草高寒草甸的平均返青期主要集中在每年的第120—140天,并且呈逐年提前趋势,30年来提前了7天。 相似文献
36.
The Three-River Headwaters Region(TRHR), which is the source area of the Yangtze River, Yellow River, and Lancang River, is of key importance to the ecological security of China. Because of climate changes and human activities, ecological degradation occurred in this region. Therefore, "The nature reserve of Three-River Source Regions" was established, and "The project of ecological protection and construction for the Three-River Headwaters Nature Reserve" was implemented by the Chinese government. This study, based on MODIS-NDVI and climate data, aims to analyze the spatiotemporal changes in vegetation coverage and its driving factors in the TRHR between 2000 and 2011, from three dimensions. Linear regression, Hurst index analysis, and partial correlation analysis were employed. The results showed the following:(1) In the past 12 years(2000–2011), the NDVI of the study area increased, with a linear tendency being 1.2%/10a, of which the Yangtze and Yellow River source regions presented an increasing trend, while the Lancang River source region showed a decreasing trend.(2) Vegetation coverage presented an obvious spatial difference in the TRHR, and the NDVI frequency was featured by a bimodal structure.(3) The area with improved vegetation coverage was larger than the degraded area, being 64.06% and 35.94%, respectively during the study period, and presented an increasing trend in the north and a decreasing trend in the south.(4) The reverse characteristics of vegetation coverage change are significant. In the future, degradation trends will be mainly found in the Yangtze River Basin and to the north of the Yellow River, while areas with improving trends are mainly distributed in the Lancang River Basin.(5) The response of vegetation coverage to precipitation and potential evapotranspiration has a time lag, while there is no such lag in the case of temperature.(6) The increased vegetation coverage is mainly attributed to the warm-wet climate change and the implementation of the ecological protection project. 相似文献
37.
38.
基于DMSP/OLS灯光数据的快速城市化过程的生态效应评价研究——以环渤海城市群地区为例 总被引:2,自引:0,他引:2
快速有效地评估城市化过程带来的生态环境后果,对于优化城市土地利用格局、降低和防范城市生态环境风险,非常必要。因此本文综合利用DMSP/OLS夜间灯光数据和SPOT/VGT时间序列数据等多源遥感信息,以NDVI与时间的积分值来表征一定时间段内的植被初级生产力,探讨了环渤海城市群地区城市化过程对植被初级生产力的季节性变化影响。发现:(1)研究区全年总的平均初级生产力总体表现为城市地区低于非城市地区的特征。(2)研究区平均初级生产力一般是在8月份最高,而在1月份最低;同时,一个生长季内,平均初级生产力总体呈现为4-11月城市地区低于非城市地区,而12月到次年3月则是城市地区要高于非城市地区的趋势,但这种趋势在各土地覆盖类型间也存在很大的不同。(3)研究区全年总的平均初级生产力,城市地区NDVI为110.23d/km^2,而非城市地区为123.94d/km^2,两者相差13.71d/km^2,即城市化过程已经在一定程度上减弱了研究区的植被初级生产力。 相似文献
39.
遥感图象区域面积的计算潘耀忠(内蒙古气象灾害监测服务中心)1前言用计算机分析遥感图象时,常需计算某一不规则连通区域(如湖泊、岛屿等)的面积。计算方法之一是开窗将感兴趣区域(假设为某一湖泊)全部包含于此窗中,再依次检查此窗口内的每一象素点,判别其是否具... 相似文献
40.