D-S Algorithm Based on SCO for Matching Fragments
Jia Ze Sun, Guo Hua Geng, Xiao Ying Pan · Applied Mechanics and Materials · 2013
The Dempster-Shafer (D-S) evidence theory is an effective method for uncertain information fusion. Because multiple evidences from different sources of different importance or reliability in the reassembling fractured 3D objects are not equally important when they are combined. This paper presents a social cognitive optimization algorithm (SCO) to generate optimal evidence weight values based on historical training data. In the algorithm, a constrained nonlinear optimized model is established, which is solved by SCO. Compared with the two methods, optimization weight D-S proves more effective than the traditional D-S.