Multispectral remote sensing image classification method and system

Abstract

The invention discloses a multispectral remote sensing image classification method and system. The classification method comprises steps of adopting a decimal coding method to compress multiple wavebands; converting the gray values of a plurality of wave bands into a decimal coded value, wherein the number of coding bits and the number of wave bands are the same; carrying out clustering classification on all decimal codes by adopting a minimum distance method or a spectral included angle method; obtaining reference decimal codes of the clustering center, calculating category attributes corresponding to all the decimal codes and the reference decimal codes according to a minimum distance method or a spectral included angle method, and finally, realizing automatic classification of the wholemultispectral remote sensing image in an indexing manner. Compared with an existing classification method, the multispectral remote sensing image classification method has the advantages of being higher in precision and faster in processing speed.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to a multispectral remote sensing image classification method and system. Background technique [0002] Remote sensing image classification methods are divided into two categories: unsupervised classification and supervised classification. [0003] Supervised classification is a classification method that requires learning and training. First, select a representative training site from the research area as a sample, select the characteristic parameters according to the samples provided by the known training area, establish a discriminant function, and identify non-samples based on the characteristics of the sample category. The category to which the cell belongs. Specific methods include maximum likelihood method, artificial neural network method and so on. [0004] Unsupervised classification does not require manual sampling of ground sample point data. Under the conditi...

AI Technical Summary

Problems solved by technology

[0080] Aiming at the present situation of the existing non-supervised classification method with many manual interventions, low classification accuracy, and slow processing speed, the present invention proposes a multi-spectral remote sensing method based on decimal code compression and multi-spectral remote sensing image indexing with reference to the basic principle of human eye recognition of ground objects. A fast classification method and system for images, and quantitative comparison with the classification results of the traditional ISODATA method and K-mean method, to obtain a fast and high-precision automatic classification method and system for multi-spectral remote sensing images

Images