论文部分内容阅读
提出一种极限学习机(ELM)和支持向量机(SVM)相融合的遥感图像分类模式.选取ELM为基础分类器,以SVM来修正改善分类效率.仿真实验结果表明,该算法不仅具有较高的分类精度,而且消除一些训练样本标签对分类的负面影响.结合ALOS/PALSAR、PSM图像与SVM、ANN(Artificial Neural Network)方法进行对比分析,发现该方法鲁棒性较好.
This paper proposes a remote sensing image classification model based on ELM and SVM. The ELM-based classifier is chosen to improve the classification efficiency by using SVM. The simulation results show that the algorithm not only has higher The classification accuracy of some training samples was eliminated, and the classification of ALS / PALSAR and PSM images was compared with SVM and Artificial Neural Network (ANN). The results show that the proposed method is robust.