Intelligence image fusion recognition algorithm based on Dempster-Shafer evidence reasoning
Yongmei Cheng, Quan Pan, Hongcai Zhang, Gang Wang · 2004
Three dimension object recognition is a basic problem of computer vision field. There were uncertainness, uncompleteness and unclarity in acquirement data as invariant moment theory was applied to recognize three dimension image target, therefore recognition efficiency could not meet the needs of actual system. By combining Dempster-Shafer evidence reasoning with invariant moment theory, intelligence image fusion recognition algorithm was present. A basic belief assignment function was constructed, and a classify rule was built. Dempster combination rule and the absorptive method were used for recognizing three dimension airplane image. The result of simulation indicated that absorptive method had higher recognition efficiency than Dempster combination rule.