摘要
针对航空遥感高光谱图像大数据量快速分析的需求,本文提出了一种基于支持向量机(SVM)的航空高光谱赤潮监测方法。首先,利用对数残差法(LRC)归一化高光谱数据。然后,通过RPCL(Rival Penalized Competitive Learning)聚类分析划分训练样本空间从而形成训练样本集,并以支持向量机(SVM)作为识别器,试验结果证明了该方法的有效性。
This paper proposes a red tide monitoring method based on SVM. Firstly the Log Residual Correction (LRC) is used to normalize the data. Then clustering analysis is adopted to select and form the training samples. The discriminator is composed of support vector machines (SVM). The experiments show that this method can complete the red tide monitoring quickly and effectively.
出处
《微计算机信息》
北大核心
2008年第21期225-227,共3页
Control & Automation
关键词
高光谱数据
赤潮监测
支持向量机
hyper-spectral data
red tide monitoring
support vector machine