Author(s)

Liu Shangzheng, Liu Bin

Corresponding Author

Liu Shangzheng

Abstract

The naive Bayesian classification algorithm of infrared remote sensing image is affected by the similarity of uniform eigenvectors of different categories, which leads to the decline of classification accuracy. Therefore, the naive Bayesian classification algorithm of infrared remote sensing image with elastic model is proposed. The spring elongation distance and the elastic coefficient decibel analog sample size and classification criteria are used to exclude the Naive Bayesian classification algorithm from being affected by the conditional independent hypothesis, avoiding noise interference, and thus achieving the goal of improving the accuracy of the classification algorithm. The experiments on infrared remote sensing images show that the naive Bayesian classification algorithm based on the elastic model has strong operability and can improve the classification accuracy.

Keywords

Infrared remote sensing image; Elastic model; -Naive Bayesian algorithm