In order to solve the problem of the slow decision speed of the Support Vector Machine (SVM) due to not sparse enough in the application of Inertial Confinement Fusion (ICF) experimental targets recognition, Relevance Vector Machine (RVM) is proposed to recognize the ICF experimental targets. A multi-classification RVM based on the binary tree is designed. For getting a more reasonable binary tree structure, both the class distance and class distribution are considered in the construction of multi-classification RVM. Experiments show that the comparable classification accuracy of RVM and SVM is all square, as well as much faster decision speed due to higher sparsity. This multi-class recognition algorithm exhibits superiority in mixed classification accuracy comparing to the traditional methods such as ‘One Against One’, ‘One Against Rest ’,‘Directed Acyclic Graph’ and ‘Binary Tree Based On Class Distance’.