A multi-source information fusion method for conflicting evidence

By calculating the differences and rejection between evidence, the weight of the evidence is re-assigned, and the D-S evidence theory is used on the basis of new evidence, solving the problem of conflict of sensor signal evidence, and improving the accuracy and reliability of the integration of multi-source information.

CN114444585BActive Publication Date: 2025-01-28BEIJING INST OF TECH
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
CN202210035020.7
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-01-13
Publication Date
2025-01-28
Estimated Expiration
2042-01-13

AI Technical Summary

Technical Problem

In practical applications, due to sensor failure or external interference, the evidence of different sensor signals is highly conflicting. D-S evidence theory is difficult to achieve effective fusion of multi-source information, resulting in the target recognition results that are contrary to common sense.

Method used

By calculating the degree of differences and rejection between the evidence, the weight of each evidence is re-assigned, the new evidence after the weighted average is obtained, and the D-S evidence theory is used on the basis of the new evidence to resolve the impact of conflict evidence.

Benefits of technology

The effective treatment of conflict evidence has been realized, the accuracy and reliability of the integration of multi -source information is improved, and the target recognition results consistent with common sense are obtained.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114444585B_ABST
    Figure CN114444585B_ABST
Patent Text Reader

Abstract

The present invention relates to a multi-source information fusion method for conflicting evidence, and belongs to the technical field of target recognition. The multi-source information fusion method for conflicting evidence of the present invention specifically includes the following steps: (1) determining the recognition framework of target recognition; (2) obtaining the basic probability distribution function of the corresponding evidence according to the sensor measurement information; (3) obtaining the degree of rejection of each piece of evidence to other evidence based on the weighted idea, assigning a weight to each piece of evidence based on the rejection degree, and further obtaining new evidence obtained by weighted average of all evidence after weight assignment; (4) applying D-S evidence theory to fuse the obtained new evidence to obtain the final fusion result; (5) completing the recognition task of the target object according to the decision criterion. The present invention takes into account the importance of different pieces of evidence, and accordingly reallocates weights to different pieces of evidence, so that the fusion result is more accurate and the influence of conflicting evidence on the fusion result is effectively resolved.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • A multi-source data fusion identification method based on an evidence fuzzy factor

    CN109766933A

  • Rock mass instability stage prediction method and device based on multi-feature fusion

    CN110457757A

  • Gas-fired boiler fault diagnosis method and device based on data fusion

    CN113159138A