论文部分内容阅读
针对支持向量机遥感图像样本较大,图像中可能存在噪音、孤立点、交叉点等干扰,从而降低了支持向量机的准确率与泛化能力。针对以上情况,本文提出支持向量机与主动学习相结合的遥感图像分类方法,在训练过程中,利用主动学习策略主动选择最佳的样本进行标记类别,加入训练样本集,重新训练SVM分类器,重复此过程直到满足某些要求。此方法可以得到很好的分类效果,并减少了训练样本量。通过对遥感图像的分类试验证明了该分类方法的有效性。
For support vector machine remote sensing image larger sample, the image may exist noise, isolated points, intersections and other interference, thus reducing the accuracy and generalization ability of support vector machines. In view of the above, this paper proposes a remote sensing image classification method based on SVM and active learning. In the process of training, the active learning strategy is used to select the best samples for tag classification, add the training sample set, retrain the SVM classifier, Repeat this process until some of the requirements are met. This method can get good classification results, and reduce the training sample size. The classification experiment on remote sensing images proves the validity of this classification method.