荒漠化地区植被因子的定量反演方法

Methods of Quantitative Retrieving of Vegetation Factors for Desertification Area

采用国产高光谱分辨率成像光谱仪(OMIS-Ⅰ)系统数据对荒漠化评价的植被因子(植被盖度、生物量)进行了定量反演。通过建立以像元为单位的定量化遥感信息模型,获得荒漠化地区植被因子的分布图。结果表明,用高光谱数据定量反演荒漠化地区植被生物量和盖度是比较可靠的。当反演区域内灌木和草地同时存在时,多项式模型的精度要明显高于线性模型;当植被类型单一时,模型即为较高精度的线性模型。植被因子的定量反演与植被类型有关。

Quantitative Remote Sensing Information Model （RSIM） was studied to retrieve the vegetation factors（ vegetation cover and biomass） for desertification assessment by using the data of state-produced hyperspectral resolution imaging spectrometer（ OMIS- I ） , and the corresponding vegetation factors recoding maps based on pixel of the visual interpretation were obtained. Result shows that it is reliable to retrieve vegetation cover and biomass quantitatively by the data of hyperspectral resolution imaging spectrometer. When there are both shrub and grassland in the retrieved region, the precision of the polynomial model is much higher than that of the linear model. However, when the vegetation type is simplified, the linear model has a higher precision. The quantitative retrieving of vegetation factors are related to vegetation type.