论文部分内容阅读
精确提取作物种植面积一直是农业遥感关注的主要问题之一。综合运用低分辨率的时相变化特征和中分辨率的光谱特征,提出一种夏玉米识别方法。首先基于MODIS NDVI时间序列曲线,分析夏玉米在时相变化上的识别特征,构建识别模型。夏玉米纯像元利用识别模型识别,而耕地和非耕地类型的植被产生的混合像元,则基于像元分解办法获取耕地组分的NDVI时序特征,再利用识别模型判定,然后结合土地利用数据根据空间关系得到中分辨率结果;玉米与其他作物的混合像元则利用中分辨率尺度光谱差异加以区分。研究结果表明,在伊洛河流域主要农业区,识别精度达到90.33%,为作物类型识别提供了新的思路。
Accurately extracting crop acreage has been one of the main concerns of agricultural remote sensing. By comprehensively using the characteristics of low-resolution temporal variation and spectral characteristics of mid-resolution, a summer maize identification method is proposed. First, based on MODIS NDVI time series curve, the recognition characteristics of summer maize in time-phase change were analyzed to build a recognition model. Summer maize pure pixels are identified by using recognition model, while the mixed pixels produced by cultivated and non-cultivated types of vegetation are used to obtain NDVI temporal characteristics of arable land components based on the pixel decomposition method, and then the identification model is used to determine, and then the land use data The intermediate resolution results are obtained from the spatial relationships; the mixed pixels of maize and other crops are distinguished by the mid-resolution scale spectral differences. The results show that in the major agricultural areas of Yiluo River basin, the recognition accuracy reaches 90.33%, which provides a new idea for crop type identification.