1 / 20

Paper Reading

Paper Reading. Author: Zuchao Wang, Min Lu, Xiaoru Yuan, Junping Zhang, Huub van de Wetering. Visual Traffic Jam Analysis Based on Trajectory Data. A programmer’s perspective. Explainer: Yabo Su. Outline. Paper information Paper structure Work flow and its details Proprocessing part

umed
Télécharger la présentation

Paper Reading

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Paper Reading Author: Zuchao Wang, Min Lu, Xiaoru Yuan, Junping Zhang, Huub van de Wetering Visual Traffic Jam Analysis Based on Trajectory Data A programmer’s perspective Explainer: Yabo Su

  2. Outline • Paper information • Paper structure • Work flow and its details • Proprocessing part • Visualization part

  3. Paper Information • Conference Paper • IEEE Transactions on Visualization and Computer Graphics (VAST'13) • Key Point • Traffic jam propagation

  4. Before we moving forward • Key problem to be solved: • How traffic jam propagates from one road to other road? • Questions: • What factors make this problem interesting? • How would you solve it qualitatively if you are provided historical taxi trajectory data and road network data? • How you define traffic jam using the information in the datasets? • How you detect traffic jam accroding to your defination? • How you visualize the traffic jam you have detected?

  5. Paper structure • 1 INTRODUCTION • 2 RELATEDWORK • 2.1 Traffic Event Detection • 2.2 Traffic Visualization • 2.3 Propagation Graph Visualization • 3 OVERVIEW • 3.1 Design Requirement • 3.2 Description of Input Data • 3.3 Traffic Jam Data Model • 3.4 Work Flow • 4 PREPROCESSING • 5 VISUALIZATION DESIGN • 6 VISUALIZATION RESULTS ANDCASE STUDY • 7 DISCUSSION • 8 CONCLUSION ANDFUTUREWORK • APPENDICES Why we build this system How the system is built How you use the system to get information we want to privide to you?

  6. Workflow and details

  7. Objective: • Extract all drivable ways • Ensure all roads are connected How to? Premise: You should be familiar with the road network data(OpenStreetMap). How to next? Easy, no particular algrothms!

  8. Objective: • Remove the useless trajectory data What kinds of trajectory data should be removed? Actions in the paper: Unrealistic coordinates Duplicated Time Stamp High Speed Long Distance Long Time Parking Waiting for Passenger Tiny Trajectory One point constraint Two points constraint multi points constraint A stop dection algorithm[37]

  9. Objective: • Map trajectory points to the road network How to? Actions in the paper: It’s a complicated probem. ST-matching algorithm[30] Result: <dWay, List<trajectory point>>

  10. Objective: • Map trajectory points to the road network How to? Actions in the paper: Removing the taxis with exeptionally high speed road speed = an average of the average speeds on the remaining taxis Record “support” Result: <dWay, road speed, support>

  11. How to? • Objective: • Detect traffic jam events Actions in the paper: Sort all valid speeds for a dWay in ascending Pick the speed value at the percentage F% position With a valid speed less than percentage C% of the free flow speed, is said to have a low speed Result: List<traffic jam event> traffic jam event - <d, t0, t1>

  12. How to? • Objective: • Find the events propagation Actions in the paper: Assumption: e1->e2 <=> e1.t0<= e2.t0<=e1.t1 && e1.d->e2.d 2. Use the STOTree algorithm [29] construct the progation graphs Result: List<progation graph> progation graph - <V, E> direct, acyclic number of events, time span, total distance

  13. Visualization design • 1. Pixel Based Road Speed View • 2. Graph List View • 3. Graph Projection View • 4. Spatial View • 5. Multi-faceted Filter View

  14. 1. Pixel Based Road Speed View

  15. 2. Graph List View

  16. 3. Graph Projection View

  17. 4. Spatial View

  18. 5. Multi-faceted Filter View

  19. Overview again

  20. Thanks

More Related