K-均值聚类算法在多波束底质分类中的应用

针对海底采样点较少时,监督学习训练分类模型困难的问题,研究无监督学习的K-均值聚类分析算法在多波束海底底质分类中的应用。在探讨K-均值聚类分析算法原理的基础上,构建海底底质分类器,针对分类器需预先输入分类结果种类(K值)这一问题,提出了基于底质采样点和分类效果连续性为原则的K值确定方法。实验结果表明:基于K-均值聚类分析算法的海底底质分类器能较好的实现海底底质类型的自动划分,适用于海量多波束底质特征参数的分类。

The Application of K-means Clustering Analysis Algorithm in Multibeam Seafloor Classification

For lack of bottom sampling points,the classification model with supervised learning training is difficult.Therefore the K-means Clustering Analysis Algorithm which is unsupervised learning is studied in multibeam seafloor classification.Based on the theory of the K-means Clustering Analysis Algorithm,the seafloor classifier is built.Due to the classification results must be input in the classifier in advance,the K-value methods are discussed,which is based on the role of the continuity of the bottom sampling points and the classification quality.The experiment result shows that the types of the bottom sediments can be classified automatically with the classifier which is based on the K-means Clustering Analysis Algorithm,and the method can be applied in the classification of the mass multibeam bottom feature parameters.