In synergy recognition, the fusion of multiple sources has to be considered to increase the reliability of target recognition. In classification fusion problem, it is considered that the classifiers have different weights to solve the problem that different classifiers may have different classification quality. DS theory is expert at characterizing and settling uncertain information. Thus, DS theory is used to optimize the classifier weights. Classifier weighting can improve the accuracy of the classifiers. However, the accuracy of the classification results of common classifier for different objects may also be different. Thus, for each object, we comprehensively use the evidence distance and conflict coefficient to measure the degree of inconsistency between the evidence, so as to determine the importance coefficient of each object. Then, through the evidential discounting operation, the basic belief assignment is discounted by considering the weights of objects to reduce the impact of high conflict, and the discounted basic belief assignments are finally combined by DS theory for making the class decision. Real data sets are used to evaluate this method, and results show that the proposed method can efficiently improve the classification accuracy.
Refined Weighted Fusion Recognition Method for High Conflict Information
Lect. Notes Electrical Eng.
International Conference on Autonomous Unmanned Systems ; 2021 ; Changsha, China September 24, 2021 - September 26, 2021
Proceedings of 2021 International Conference on Autonomous Unmanned Systems (ICAUS 2021) ; Kapitel : 314 ; 3204-3210
2022-03-18
7 pages
Aufsatz/Kapitel (Buch)
Elektronische Ressource
Englisch
TESTED - Ford Fusion - Refined US version of new Mondeo
Online Contents | 2012
Conflict recognition in air traffic control
Tema Archiv | 1996
|