论文部分内容阅读
使用新型遥感数据——“北京一号”小卫星数据,采用BP神经网络法对密云水库流域内的植被覆盖进行反演,并将结果与传统回归分析法和NDVI像元二分法进行比较。结果表明:在山区植被信息遥感反演算法中,神经网络方法以其对非线性过程的精确模拟而具有比传统算法更高的精度,尤其对于遥感反演算法难度较大的山区植被覆盖信息提取效果较好。
Using the new remote sensing data - Beijing No. 1 satellite data, the BP neural network method was used to invert vegetation cover in the Miyun Reservoir watershed and the results were compared with the traditional regression analysis and the NDVI pixel dichotomy . The results show that in the remote sensing inversion algorithm of mountain vegetation information, the neural network method has higher precision than the traditional algorithm for its accurate simulation of nonlinear process, especially for remote sensing retrieval algorithm which is difficult to extract mountain vegetation cover information The effect is better.