Research On Semantic Method Of Scientific Examination Track Based On Residence Area Division

Polar scientific expedition is an important way to promote the cause of scientific expedition in China.In recent years,China has made rapid progress in the field of polar expedition.As an indispensable expedition vessel for polar scientific examination in China,the “Xuelong” is a strong support for scientists to go to polar regions for scientific examination.During the decades of “Xuelong” voyage,due to the long distance of the polar scientific examination and the bad environment,the cycle often lasts for several months.Every voyage will produce tens of thousands or even hundreds of thousands of track points,and accumulate a large number of track data.A large amount of track data.As one of the important results of the polar expedition,the scientific track data obtained a sequence of continuous time and space points according to a certain acquisition frequency.Each track point records the time,position,heading,speed and other movements of the scientific research ship at that point.Information and historical track data contain great research value and can provide important reference for subsequent navigation.They are concerned by the majority of polar research scholars.However,the quality of the original track data is not high,and it has the characteristics of long period,different sampling intervals and uneven information density,which brings difficulties to data analysis and association information mining in the process of track semantics.In order to extract effective track segments and excavate potential information of track data,existing track studies have focused on segmentation and semantics,and some progress has been made.However,there are still some problems:(1)In the aspect of track segmentation,speed thresholds and clustering methods have been partitioned by factors such as frequent fluctuation of speed,density of data sampling,overlapping of spatial data and so on.There are limitations in the process.(2)In the research of track semantics,the existing research on track semantics is mainly based on the realization of stay.However,due to the limitation of ambiguous division of sea area and the inherent characteristics of track data,the results of semantics are not ideal,and need to be further studied.(3)There is no systematic overall process for the Semantics Research of the scientific examination track in the existing track research.In view of the above problems,this paper proposes a systematic process and framework for the scientific study track semantics from the perspectives of track data characteristics,speed state changes and track semantics.The work content and innovations mainly include three aspects:(1)Proposed the scientific research on the track trajectory semantics.Firstly,the problem of noise data,data missing and data duplication in the scientific track data is combined with the existing methods to preprocess the scientific track data,eliminating noise and duplicate data.(2)Aiming at the problem of frequent fluctuation of the speed of scientific expedition track,a track segmentation method based on track time domain difference(TDD)is proposed.The method is based on the time domain to differentially process the speed of the ship,which effectively reduces the influence of the frequent fluctuation of the speed on the segmentation result.At the same time,considering the ambiguity of the differential step size of the method in the track processing process,this paper will differentiate the back navigation.The path loss and the speed fluctuation amplitude of the track are normalized,and the dynamic determination method of the track difference time step is proposed,and the track is segmented by the rate threshold.(3)Based on the definition of trajectory semantics,according to the characteristics of the scientific research track stay area,the scientific test track semantics is redefined: the stay area range R,the stay area enters the world tin,the stay area departure time tout,and the multi-source information flag Fmsi,The scientific research method of the scientific research track semantics is proposed by combining the track data preprocessing and the TDD track segmentation method to realize the semantics of the scientific test track.Taking the 29 th Antarctic expedition track data as an example,this method is compared with the classical DBSCAN track segmentation method.The experimental results show that the proposed method can effectively reduce the problem of excessive segmentation during track segmentation.The results are better in terms of segmentation accuracy and time efficiency.After that,the data of the semantic expedition track is also represented by the track data example,which proves the feasibility of the semantic method and enriches the data information of the track.Finally,the hotspot area function is displayed as the system application and implementation of the track semantics method.