Research On Trajectory Similarity Analysis Of Fishing Boats Based On AIS

As an important marine fishery production tool,the fishing boat's running trajectory data contains a large amount of spatial and temporal distribution information.Accurate analysis of the fishing boat's trajectory information can excavate the fishing boat's sea-going laws and operating behavior patterns,which is of great significance for accurately grasping the operation behavior of fishing vessels,discovering abnormal behavior of fishing vessels,and standardizing fishery law enforcement.At present,a large number of domestic and foreign scholars have used the time,latitude and longitude,heading,speed and other attributes of the trajectory points in the AIS trajectory data.Through the establishment of mathematical models and research algorithms to process trajectory data,to process the trajectory data through the establishment of mathematical models and research algorithms.However,most of the research objects are commercial ships,and the research in the field of fishing boats is not deep.Therefore,this article takes fishing boats as the research object,and conducts research and analysis based on a large number of fishing boat AIS trajectory data.According to the data characteristics of the fishing boat trajectory,carry out relevant preprocessing work on the AIS trajectory data of the fishing boat,and then carry out similarity analysis work on it.The specific work can be summarized as the following two points:1)A fishing boat AIS trajectory data compression algorithm based on improved sliding window is proposed:In view of the low quality of the AIS trajectory data of fishing vessels and the analysis of the trajectory of fishing vessels,it is necessary to retain more feature points.Based on the Sliding window algorithm,the driving feature points are determined by using the latitude and longitude change status trend between adjacent track points,compress high-density stagnation points while retaining timing information,use the sampling method to keep straight intermediate points,keep as many track feature points as possible.Experiments show that the fishing boat AIS trajectory data compression algorithm based on the improved sliding window controls the compression rate to below 94%,which keeps the trajectory feature points of the fishing boat well,and also avoids the trajectory distortion of the Sliding Window algorithm,which is more conducive to subsequent Analysis of fishing boat operations.2)A similarity measurement algorithm for fishing boat AIS trajectory based on piecewise dynamic programming is proposed: For the similarity measurement of AIS trajectory data of fishing boats,there is a problem that the trajectories are not equal in length and there is a time drift between the compared trajectories.This paper combines the characteristics of the sampling rate of trajectory points of different trajectories is different,and proposes a segmented improvement idea on the known dynamic time planning algorithm,put forward the improved idea of segmentation on the known dynamic time planning algorithm,the point-to-segmentdistance calculation is used to reduce the influence of the trajectory sampling method on the sensitivity,and then the angular distance is introduced into the segment-to-segment distance calculation to improve the accuracy of the similarity measurement.Experiments show that in the two clustering comparison experiments,the accuracy of the similarity measurement algorithm for fishing boat AIS trajectory based on piecewise dynamic programming is 15.5%and 5.3% higher than the Hausdorff algorithm and DTW algorithm,respectively,which can achieve an ideal similarity measurement result.