Driver fixation point clustering method based on density clustering method and morphology clustering method

Abstract

The invention provides a driver fixation point clustering method based on a density clustering method and a morphology clustering method, and belongs to the field of typical density clustering methods and mathematic morphology clustering methods. The driver fixation point clustering method includes the steps of putting forward a density method and mathematic morphology method combined self-adaption DBSCAN-MMC method, applying the method to driver fixation point clustering, setting the value of the Eps through fixation point structure parameters, obtaining an initial point set of MMC clusters through the DBSCAN, determining the number of the clusters, reducing outliers produced through DBSCAN clusters through the self-adaption MMC clusters, and completing clustering oriented to driver fixation areas. According to the method, the advantages of irregular shape clustering of the DBSCAN and the MMC are fully used, the defects of the two clustering methods are overcome, the clustering effect is superior to the clustering effect of the conventional DBSCAN clustering method and the conventional MMC clustering method when the driver fixation areas are divided, and the driver fixation clustering quality is improved.

Description

technical field [0001] The present invention belongs to the typical density clustering method DBSCAN (DensityBasedSpatialClustering ofApplicationswithNoise), the clustering method field of mathematical morphology clustering method (Mathematical Morphology Clustering, MMC), particularly relates to a kind of driver visual behavior field and classification in traffic engineering The field of clustering in science. Background technique [0002] Using the clustering method to divide the driver's gaze area can overcome the shortcomings of the traditional gaze area division, which is highly subjective, and help to find the driver's visual law, thereby improving the accuracy of driver state monitoring and driving behavior prediction. However, the distribution of the driver's gaze points is irregular and discrete, and the conventional distance-based clustering method has defects such as a good clustering effect on "quasi-circular" data and a relatively rigid cluster boundary. Theref...

AI Technical Summary

Problems solved by technology

[0006] The technical problem to be solved by the present invention is: for the conventional distance-based clustering method, the clustering effect of irregularly distributed and relatively discrete non-"quasi-circular" driver gaze point data is poor, and the clustering boundary is relatively rigid. However, the clustering effect of typical density clustering DBSCAN, which can generate point clusters of arbitrary shapes, is greatly affected by parameter values, and mathematical morphology clustering requires a lot of manual intervention. A DBSCAN- MMC clustering method for automatic and efficient clustering of driver gaze points

Images