2019_programme: THREATENING DEGREE ASSESSMENT METHOD OF SONAR TARGETS BASED ON MULTI-SOURCE INFORMATION FUSION



  • Session: 04. Towards Automatic Target Recognition. Detection, Classification and Modelling
    Organiser(s): Groen Johannes, Jans Wolfgang, Pailhas Yan, Myers Vincent
  • Lecture: THREATENING DEGREE ASSESSMENT METHOD OF SONAR TARGETS BASED ON MULTI-SOURCE INFORMATION FUSION
    Paper ID: 791
    Author(s): Zhou Bin, Song Xuejing, Wang Qing, Chen Yuechao
    Presenter: Zhou Bin
    Presentation type: oral
    Abstract: Multi-source information fusion can expand the detection range of a sonar system and improve the reliability and robustness of target recognition. To solve the problems of large transmission bandwidth, uncertain priori probability of target threatening degree discrimination, and inability to perform fusion processing within the framework of probability theory in the sonar target association using full spectrum feature information, a threatening degree assessment method for sonar targets based on multi-source information fusion is proposed in this paper. The method combines the target information output from low-frequency array, high-frequency array and non-acoustic sensor, to perform threatening degree assessment of low-frequency sonar target through multi-source information preprocessing, target track association, association information fusion and target classification. For target track association, an improved gray association algorithm is proposed, which can distinguish two different targets with the same track change trend but far away from each other. In the absence of a priori probability of threatening target discrimination, a multi-source information fusion method based on Dempster-Shafer evidence theory is proposed to solve the problem that probability theory cannot be effectively applied to information fusion processing, so as to eliminate non-threatening targets. Finally the Template matching technique is used to further improve the accuracy of target threatening degree assessment. The proposed method is verified by sea trial data of multi-source information, including low-frequency array data, high-frequency array data and automatic identification system(AIS) data. This work can meet the needs of low-frequency sonar target threatening degree assessment.
      Download the full paper
  • Corresponding author: Mr Zhou Bin
    Affiliation: 1. Science and Technology on Sonar Laboratory´╝Ť2. Hangzhou Applied Acoustics Research Institute
    Country: China
    e-mail: