Hyperspectral image spectral space classification method based on class characteristic iterative random sampling

Abstract

The invention discloses a hyperspectral image spectral space classification method based on class characteristic iterative random sampling. The method comprises the following steps: calculating classcharacteristic criteria of each class according to hyperspectral image data and ground object class label information; iteratively calculating the number of training samples which should be allocatedto each category during classification according to a category feature criterion; calculating an average spectral characteristic and an autocorrelation matrix of a target ground object according to the number of the distributed training samples of each type and the parameter information of the target ground object sample set; calculating abundance information corresponding to each pixel in the hyperspectral image for the ground object through a constraint energy minimization method according to the average spectral characteristic of each ground object and the inverse matrix of the autocorrelation matrix; and calculating a similarity coefficient of two adjacent iterative classification results according to the classification image information, judging whether the similarity coefficient meets an iterative classification cut-off condition, and carrying out iterative classification on the fused and updated data.

Description

technical field [0001] The invention relates to a hyperspectral image classification method based on class feature iterative random sampling and spatial information feedback, in particular to a hyperspectral image spectral space classification method based on class feature iterative random sampling. Background technique [0002] Hyperspectral remote sensing images include hundreds of continuous narrow-spectrum spectral bands, so a large number of spectral details can be used to accurately determine the category of ground objects. Therefore, it has been widely used in land use change and cover, disaster monitoring, geological assessment, agricultural and forestry survey, etc. Hyperspectral image classification is to classify each pixel in the hyperspectral image. Compared with conventional remote sensing, the spectral resolution of hyperspectral images is greatly improved while retaining high spatial resolution. The ability to describe the details of similar features and to ...

AI Technical Summary

Problems solved by technology

However, when the number of known class samples is small, the classification effect of ICEM is poor, so the implementation of ICEM requires all the class label information of known samples to calculate the spectral mean of each target class as the target spectrum

However, in actual situations, obtaining labels for hyperspectral data samples is time-consuming, labor-intensive, and costly, so that it is very difficult to obtain sample label information for all categories.

Images