The Combined Use of the Evidence Theory and Fuzzy Logic for Improving Multimodal Nondestructive Testing Systems

A data fusion approach for improving weld inspection by multimodal nondestructive testing (NDT) is investigated. The method relies on the combined use of evidence theory and fuzzy logic. In this framework, X-ray and ultrasonic inspection is modeled using fuzzy sets and mass function in evidence theory, and objects corresponding to X-ray and ultrasonic testing are fused at the level of features. The results show that the fusion improves significantly the reliability of defect detection and the defect sizing.