Classifying Hyperspectral Data Using Support Vector Machine Conditional Random Field

With there problems at hands,an improved random field,support vector machine conditional random field(SVM-CRF) was proposed.It uses SVM as its unary potential,combining the merits of SVM and CRF.Experiments using AVIRIS hyperspectral data as input were carried out,and SVM-CRF was analyzed extensively.Experimental results show that SVM-CRF is superior to SVM and classic CRF in terms of classification accuracies.