Hyperspectral image blur classification method and device based on related vector machine

Abstract

The invention discloses a hyperspectral image blur classification method and device based on a related vector machine. The method includes the following steps that a training sample set is determined, a sparse Bayesian classification model is used for selecting a kernel function, and a related vector machine classification forecasting model is built; aiming at the training sample set, a one-to-one method is adopted for constructing multiple classes of classifiers of the related vector machine, and classifier parameters are optimized through cross validation; the classifiers of the related vector machine are used for performing fuzzy classification on hyperspectral images. According to the hyperspectral image blur classification method based on the related vector machine (RVM), a sequence sparse Bayesian learning algorithm is adopted for improving training speed of the RVM, and aiming at the RVM classifiers, constructed through the one-to-one method, of the RVM, pairwise-coupled probability output is converted into the membership degree relative to ground feature classifications. Compared with a SVM, the RVM is simple in parameter selection and high in classification speed, mixed pixels can be identified by the utilization of the blur membership degree, and reliability of image classification is effectively improved.

Description

technical field [0001] The invention relates to the technical field of hyperspectral remote sensing, in particular to a hyperspectral image fuzzy classification method and device based on a correlation camera machine (RVM). Background technique [0002] Compared with panchromatic and multi-spectral images, hyperspectral images have rich spectral information of ground features and have great advantages in ground feature recognition. Hyperspectral image records have a wide spectral range and high spectral resolution, and can obtain fine spectral curves of ground objects, from which the radiation characteristic parameters of targets can be well extracted, making quantitative analysis of surface targets possible. Hyperspectral remote sensing has become a new and important technical means in the fields of vegetation survey, ocean remote sensing, agricultural remote sensing, environmental monitoring, and military intelligence acquisition. [0003] The imaging mechanism of hypersp...

AI Technical Summary

Problems solved by technology

[0009] SVM can effectively avoid the phenomenon of over-learning and has good generalization ability, but it has obvious deficiencies, which are mainly manifested in: ① The number of basis functions basically grows linearly with the size of the training sample set, and the model sparsity is limited; ② Prediction The results are not statistically significant, and the uncertainty of the prediction results cannot be obtained; ③The kernel function parameters and regularization coefficients need to be determined by cross-validation and other methods, which increases the amount of calculation for model training; ④The kernel function must meet the Mercer condition

Images