Intelligent Classification Method of Remote Sensing Image Based on Big Data in Spark Environment

摘要

In addition to providing interactive queries, Spark can also optimize the iteration workload. The processing of remote sensing image is also a hot topic in recent years. This paper improves the traditional image classification technology based on K-means algorithm, and proposes a new hyperspectral image clustering method combined with spectral clustering method. On the basis of feature dimensionality reduction of hyperspectral image data, K-means algorithm is used for rough clustering of images, and then spectral clustering method is used for quadratic clustering of the clustering results. Compared with the traditional K-means clustering algorithm, the simulation results show that this method can effectively improves the classification accuracy of image clustering. The comparison experiments between simulated data and real hyperspectral data show that this method has the best clustering performance compared with the single K-means or spectral clustering method.