交通事故时空可视化分析应用研究-计算机技术专业分毕业论文.docx

AbstractThe Abstract The issue of road traffic accidents becomes emerging topic along with the fast development of cities．How to the serious impact of traffic accident well to guarantee the commuting safety of residents，have brought to the forefront of governments，traffic management bureau，industries and academics．However,the challenges of recent accident analysis research remain：First，due to the heterogeneous case—by-case nature of traffic accidents and the data incorrectness problem，it is difficult for traffic management bureau to use directly such traffic big data for analytics in related works．Second，the lack of both requirement-oriented accident analysis and visualization applications poses difficulties for previous data mining and analytics technologies．The thesis aims to both fuse the multiple accident—related datasets and to utilize spatio-temporal analysis methods，in order to solve the problem that the traffic management bureau is facing．We propose accident spatio—temporal analysis and visualization method by fusing multiple features，to benefit the traffic management bureau with both accident immediate processing and efficient duty management． The main contents of this paper are：First，the related works and technologies of traffic big data，accident analysis，and visualization has been reviewed．Second，based the multiple traffic accident datasets from October 20 1 4 to March 20 1 5 in Xiamen， we conduct the data pre-process and data fusion．Third，we propose novel accident spatio·temporal analysis and visualization method by fusing multiple traffic features． Finally,we show the value of accident spatio-temporal analysis and visualization method through three real-world case studies in Xiamen．The major contributions of this paper twofold．First，the accident analysis and visualization method based spatio—temporal correlations is proposed．Second，we present accident visualization methodology fused with multiple traffic features． Traffic big data analytics and visu