Track similarity calculating method

Abstract

The invention discloses a track similarity calculating method. The track similarity calculating method includes steps of 1, collecting user space data by handholding a GPS device, and taking two tracks R and S; 2, converting longitude and latitude coordinates to a general horizontal axis Mercator projection coordinate system, and filtering the track data by a Kalman filtering algorithm; 3, combining point section distance with inter-point Euclidean distance, and calculating the improvement space distance of every point of two tracks R and S; 4, calculating the predicting distance of every point in two tracks R and S; 5, using results and intersectional angle distance of step three and step four, calculating distance of every section of two tracks R and S; 6, calculating the accumulated distance of two tracks R and S; 7, performing normalization processing on the result of step 6. The track similarity calculating method has the beneficial effect of reducing the influence of the track sampling frequency on accuracy of algorithm and improving the accuracy of track similarity algorithm.

Description

technical field [0001] The invention belongs to the field of computer applications, in particular to a method for calculating trajectory similarity. Background technique [0002] With the development of sensor technology and personal smart devices, GPS devices are widely used to track moving objects (people, vehicles, animals, etc.), and a large amount of trajectory data is continuously generated every day. Trajectory is the spatiotemporal data sequence left by a moving object in space over time. With the promotion of the Internet of Things, urban computing and other fields, the storage and analysis of spatiotemporal data has become a research hotspot in the field of database and machine learning. Through analysis These trajectory data can bring huge impetus to all walks of life, such as planning playground infrastructure construction by analyzing the movement trajectory of playground visitors, analyzing the movement patterns of outstanding football players, and analyzing th...

AI Technical Summary

Problems solved by technology

However, DTW is sensitive to the sampling frequency, and does not use the trajectory shape as an element to calculate the trajectory similarity, resulting in a less than perfect calculation accuracy.

Images