Abstract: The growing number of ships at sea and the complexity of maritime operations significantly increase the load on maritime communication links and channel occupancy, intensifying channel congestion. Maritime communication systems often operate independently, leading to inconsistent management and inefficient bandwidth allocation. IALA（The International Association of Marine Aids to Navigation and Lighthouse Authorities） warns that when maritime channel load exceeds 50%, signal transmission may experience delays due to information blockages. To protect AIS（Automatic Identification System） channel resources and ensure maritime traffic safety, I propose an intelligent identification algorithm that addresses the challenge of recognizing fishing net markers due to irregular information, which impacts navigation safety. This algorithm models the group movement characteristics of fishing net markers and combines supervised and unsupervised learning methods to accurately identify fishing net markers within large AIS data sets that include ships and fishing nets. The results show that the algorithm effectively identifies fishing net markers, achieving an accuracy rate of over 95%, enhancing control over these markers and reducing potential risks to maritime safety.