TRB 89 th Annual Meeting - PowerPoint Presentation

Slide1

TRB 89th Annual Meeting

Traffic Monitoring of Motorcycles during Special Events Using Video Detection

Dr. Neeraj K. KanhereDr. Stanley T. BirchfieldDepartment of Electrical EngineeringDr. Wayne A. Sarasua, P.E.Sara KhoeiniDepartment of Civil Engineering

College of Engineering and Science

Clemson UniversitySlide2

TRB 89th Annual Meeting

Introduction

Data from NHTSA FARS indicates disturbing trends in motorcycle safety In 2006, motorcycle rider fatalities increased for the ninth consecutive year.During this period, fatalities more than doubledSignificantly outpaced motorcycle registrationSlide3

Traffic data collection and motorcycles

In June 15, 2008 FHWA began requiring mandatory reporting of motorcycle travel as part of HPMS

Need VMT data as well as crash data to assess motorcycle safetyIn September, 2008, an HPMS report indicated that the quality of MC data was questionable due to the inability and inconsistency of current traffic monitoring equipment.TRB 89th Annual Meeting Slide4

TRB 89th Annual Meeting

Challenges with motorcycles

Three main reasons why motorcycles are difficult to count: light axle weight low metal mass narrow footprint Historically, collection of motorcycle data has been a low priority. Many commercially available classification systems are generally unable to accurately capture motorcycle traffic. Emphasis in the past has been on detection. Slide5

Overview of this research

Significant

amount of motorcycle trafficVariety of formations Chose a motorcycle rally Myrtle Beach, SC TRB 89th Annual Meeting Evaluate a computer vision based tracking system that can count and classify motorcyclesSlide6

TRB 89th Annual Meeting

Collecting Vehicle Class Volume Data

Different types of sensors can be used to gather these data: Axle sensors Presence sensors Machine vision sensors Several manufacturers indicate their devices can detect/classify motorcycles motorcycle classification accuracy specifications not available we could not identify any validation studies

Motorcycle classification with traditional sensorsSlide7

I

ssues

with length based classificationSome cars are not much longer than the average motorcycleEuropean “city cars” are gaining popularityAverage motorcycle size is larger than ever before.Cruisers have become very popularWheel base is within 10” of many subcompactsAxle counters are especially prone to length base classification errorsTRB 89th Annual Meeting Slide8

TRB 89th Annual Meeting

Loop detector

Amongst the most reliable traffic Capable of collecting speed, volume, and classifications Several configurations depending on application Length based classification is most common Adjusting detector senstivity may lead to crosstalk with trucks in nearby lanes

Motorcycle detection and classification

Classification possible w/loop arrays

Electromagnetic profiling promisingSlide9

Motorcycle Travel Symposium

Overhead and side

non-intrusive devices Active and passive infrared, radar, and acoustic devices Capable of collecting speed, volume, and classifications Length based classification is most commonMotorcycle detection and classification Vehicle profiling is possible (e.g. vehicle contour)

Some specify >99% accuracy (scanning infrared)Slide10

TRB 89th Annual Meeting

Small footprint sensors

Magnetometers Capable of collecting speed, volume, and classifications Length based classification is most commonMotorcycle detection and classification is most promising with an array of probes

spaced at 3’ to 4’ intervals

Motorcycle detection and classification Slide11

TRB 89th Annual Meeting

Axle sensors

Most are intrusive (piezo). Some temporary (hose) Capable of collecting speed, volume, and classifications Several configurations depending on application Length based and weight base classification possible Weight base may be most promising

Motorcycle detection and classification Slide12

TRB 89th Annual Meeting

Machine Vision Sensors

Proven technology Capable of collecting speed, volume, and classifications Several commercially available systems Uses virtual detection Provides rich visual information for manual inspection

No traffic disruption for installation

and maintenance

Covers wide area with a single camera

Benefits of video detectionSlide13

Motorcycle Travel Symposium

Traditional Approach to Video Detection

Limitations of localized video detection Errors caused by occlusions Spill-over errors Problems with length based classification Cameras must be placed very high (to > 40’) to minimize errorCurrent systems use localized virtual detectors which can be prone to errors when camera placement in not ideal.Slide14

Research on motorcycle video detection

Significant recent work on tracking but very little related to motorcycle detection

Duan et al. present on-road lane change assistant that can identify motorcycles using AI including Support Vector MachinesDetection rates over 90%Chiu et al. uses an occlusion detection and segmentation method using visual length and width and helmet detection.95% recognition rate for a field study of 42 motorcyclesTRB 89th Annual Meeting Slide15

TRB 89th

Annual Meeting

Clemson’s tracking approach Tracking enables prediction of a vehicle’s location in consecutive frames. Slide16

Clemson System demo

TRB 89th Annual Meeting Slide17

Algorithm Overview

TRB 89th Annual Meeting Slide18

Simple Calibration

TRB 89th Annual Meeting Slide19

Classification

TRB 89th Annual Meeting Slide20

Classified vehicles

TRB 89th Annual Meeting Slide21

Oops…

TRB 89th Annual Meeting Slide22

Field evaluation of Clemson system

First attempt at automated motorcycle data collection at a bike rally

Literature indicated several manual effortsJamar type countersPost processing videoSturgis has been used automated counters since 1990 but only to collect total vehicle volumesTRB 89th Annual Meeting Slide23

Camera details

Pan-Tilt-Zoom

Autofocus with automatic exposure640 x 480 resolution30 frames per secondTRB 89th Annual Meeting Slide24

Data collected at 2 locations

TRB 89th Annual Meeting Slide25

Summary of Results

TRB 89th Annual Meeting

ApproachingDepartingTotalMC Actual Counts

805

684

1489

MC System Counts

784

714

1498

MC Percent of Difference

-2.61

4.38

0.6

PC and HV Actual Counts

580

598

1178

PC and HV System Counts

593

582

1175

PC and HV Percent of Difference

2.24

-2.67

-0.25

Total Actual Counts

1385

1282

2667

Total System Counts

1377

1296

2673

Total Percent of Difference

-0.57

1.09

0.22

Actual Counts

System Result

Dif (Percents)

MC

726

681

-6.19

PC and HV

333

321

-3.60

Total

1059

1002

-5.38

Myrtle Beach Site

Garden City SiteSlide26

Garden city results (both directions)Slide27

Garden city results (both directions)Slide28

Garden City results - regression analysis

PC & HVMCAll Vehicles

Slope

1.0009

0.9861

0.9925

R-Sq

1.0000

0.9998

1.0000Slide29

Myrtle Beach resultsSlide30

Myrtle Beach site video

TRB 89th Annual Meeting Slide31

Garden City site video

TRB 89th Annual Meeting Slide32

Two directions at once (speed calibrated)

TRB 89th Annual Meeting Slide33

Verifying speeds

TRB 89th Annual Meeting Slide34

TRB 89th Annual Meeting

Conclusion

Motorcycle classification within 6% of actual even in extreme conditions: Algorithm works in real time Very high volumes of motorcycles Tight formations (staggered and pairs)

Improve robustness to eliminate systematic errors

Evaluate night time/low light conditions

Augment

algorithim

with pattern-based descriptors

Future workSlide35

TRB 89th Annual Meeting

Thank you !Slide36

TRB 89th Annual Meeting

For more info please contact:

Dr. Stanley T. BirchfieldDr. Neeraj K. KanhereDepartment of Electrical Engineeringstb@clemson.edunkanher@clemson.eduDr. Wayne A. Sarasua, P.E.Department of Civil Engineering

sarasua@clemson.edu