1. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage Automated Distress Data Collection Workshop 134 Sunday, January 11, 2004, 2:00 PM - 5:30 PM, Marriott Sponsored by: A2B06 - Pavement Monitoring, Evaluation and Data Storage Sunday, January 11 1 2003 Kelvin C P Wang, University of Arkansas, Presiding University of Arkansas Email: kcw@engr.uark.edu

2. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 2 Session 1: Tutorial and State-of- the Art Tutorial on Imaging for Pavement Distress Survey, Kelvin C.P. Wang, U o f Arkansas Tutorial on Imaging for Pavement Distress Survey, Kelvin C.P. Wang, U o f Arkansas Automated Pavement Distress Collection and Processing Techniques, (NCHRP Synthesis Topic 34- 04), Ken McGhee Automated Pavement Distress Collection and Processing Techniques, (NCHRP Synthesis Topic 34- 04), Ken McGhee TxDOT Automated Pavement Surface Distress Measurement System, Carl Bertrand, Texas Department of Transportation, TxDOT Automated Pavement Surface Distress Measurement System, Carl Bertrand, Texas Department of Transportation, Application of the AASHTO Interim Cracking Protocol with An Automated System for Distress Survey, Hosin Lee, U of Iowa Application of the AASHTO Interim Cracking Protocol with An Automated System for Distress Survey, Hosin Lee, U of Iowa

3. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 3 Sessions Two: Experiences and Future Directions QC/QA Processes for LTPP Photographic Distress Data Collection and Interpretation, Gonzalo R. Rada, MACTEC Engineering & Consulting, Inc., QC/QA Processes for LTPP Photographic Distress Data Collection and Interpretation, Gonzalo R. Rada, MACTEC Engineering & Consulting, Inc., Multi-Year Automated Pavement Condition Survey in Iowa DOT, Omar Smadi, Iowa State Univeristy Multi-Year Automated Pavement Condition Survey in Iowa DOT, Omar Smadi, Iowa State Univeristy Automated Crack Detection in the State of Maryland-A Case Study, Jonathan L. Groeger, Axiom Decision Systems, Inc. Automated Crack Detection in the State of Maryland-A Case Study, Jonathan L. Groeger, Axiom Decision Systems, Inc. Future Directions of Technology Development, Kelvin C.P. Wang Future Directions of Technology Development, Kelvin C.P. Wang

4. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 4 Tutorial on Imaging for Pavement Distress Survey Kelvin C.P. Wang University o f Arkansas

5. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 5 Imaging and Automation The Workshop: Primarily Cracking The Workshop: Primarily Cracking Imaging Basics Imaging Basics  Analog and Digital  Area San and Line Scan  Resolution and Crack Size  Computer Storage Processing Automation Processing Automation

6. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 6 A Typical Analog System 35-mm Film Digital Multimedia Database Film Digitizing with a Scanner

7. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 7 Example of A LTPP Pavement Section (Digitized from 35-mm Film)

8. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 8 Example of A Pavement Section, Captured with A Digital Camera

9. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 9 Analog Versus Digital Digitization Digitization  Required for Analog Image Sources  A Digital Acquisition System: Direct Interface with Computer, or Computer Friendly Resolution: Not Directly Comparable Resolution: Not Directly Comparable  Both Format: 8-bit (Dynamic Range)  Both Formats in Recent Years: Similar Resolution  Most systems: about 2-mm pixels  New Digital Systems: Even Higher Resolution  New digital systems: 1-mm pixels

10. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 10 Area Scanning or Line Scanning (a) Area Scanning(b) Line Scanning

11. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 11 The Line Scanning Method External Sync Source Integration of 96 Simultaneous TDI Lines Speed Encoder Transverse Longitudinal TDI Line-scan Camera

12. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 12 Benefits for Line Scanning Cost-effective: up to 8192 pixels per line Cost-effective: up to 8192 pixels per line Pixel fill-factor: typically 100% to maximize sensitivity Pixel fill-factor: typically 100% to maximize sensitivity Smear-free images of fast moving objects Smear-free images of fast moving objects No need for strobing or shuttering. No need for strobing or shuttering.

13. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 13 Tradeoffs for Line Scanning Lighting Lighting  Uniform illumination in field of view  High illumination intensity Optics Optics  Accommodating large image circle diameter because of high resolution  Image distortions: aberrations in spherical optics in lenses cause

14. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 14 Dynamic Range, Crack Width and Resolution Dynamic range: 8-bit, or 256 grey-scale Dynamic range: 8-bit, or 256 grey-scale Transverse Resolution 1300-Pix2048-Pix4096-Pix Visible Crack Width 3-mm2-mm1-mm

15. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 15 Crack Width: Not A Science 8-Bit Pavement Images: 8-Bit Pavement Images:  256 levels of shades Representation of a Crack Representation of a Crack  Many levels of shades Determining Crack Width Determining Crack Width  Subjective Visual Inspection Visual Inspection  May not yield commonly agreeable crack width

16. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 16 Storage & Compression/Lane-Mile Raw ImageJPEGJPEG2000 2,048-Pix1.6 GB200 MB70 MB 4,096-Pix6.6 GB800 MB280 MB

17. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 17 Automation of Distress Survey Status Status  Fully Automated: Rutting Only  Cracking Survey: Manual, Semi-Automated, Near Fully Automated  Most Other Distress Surveys: Manual Full Automation for Cracking Survey Full Automation for Cracking Survey  Optimistic: Digital Acquisition, Storage, Computational Platforms Methodologies for Automation Methodologies for Automation  Proprietary

18. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 18 UA’s The Parallel Computing Approach for Real-Time Automated Survey GPSDMICamera Dual-CPU Acquisition Multi-CPU,Distress Analyzer CPU 1 CPU N Project Manager for Parallel Processing Expanded View of the Distress Analyzer

19. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 19 Roadway Condition Survey: Digital Highway Data Vehicle (DHDV) Developed at UA Since 1998, 2 nd Generation

20. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 20 Automated Distress Data Processing 1 Millimeter Resolution, Complete Coverage Data Collection at Highway Speed, Day or Night Automated Detection & Classification of Cracks

21. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 21 Demonstrations 4,096-Pixel (Transverse) Images 4,096-Pixel (Transverse) Images Distress Analyzer Distress Analyzer

22. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 22 Data with AASHTO Interim Protocol, 2.8-Mile

23. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 23 Data Using Universal Crack Indicator (CI), 2.8-Mile

24. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 24 Four-Pass Test with Universal Crack Indicator (CI), 2.8-Mile

25. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 25 Performance of the Current Automated System, DHDV Accuracy Accuracy  Produce crack map and geometrics of nearly all cracks shown in images  Classify vast majority of longitudinal, transverse, block, & alligator cracks  Key to High Accuracy: Image Quality Processing Speed Processing Speed  Over 60 MPH

26. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 26 Two Passes of 25-Miel Section of Pavement, # of Transverse Cracks

27. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 27 Conclusion Digital Technology Digital Technology  More Prevalent  Higher Perceived Resolution than Analog Format  Easier to Operate  A Key to High Quality Image: Illumination Automation Automation  Full Automation for Cracking & Rutting Survey: A Reality  Other Distresses: More Research

28. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 28 Automated Pavement Distress Collection and Processing Techniques NCHRP Synthesis Topic 34-04 Ken McGhee Topic consultant

29. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 29 Topic 34-04 Objective/definition Objective : Document how agencies conduct automated pavement distress data acquisition and processing Definition of automated : Distress captured by imaging or non-contact sensors is included

30. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 30 Synthesis Scope – Distresses Considered Pavement cracking - major emphasis Pavement cracking - major emphasis Pavement ride quality Pavement ride quality Pavement rutting Pavement rutting Slab faulting Slab faulting

31. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 31 Synthesis Scope – Issues Considered Contrast state-of-art and state-of- practice Contrast state-of-art and state-of- practice Procurement/contracting procedures Procurement/contracting procedures Quality control/quality assurance Quality control/quality assurance Monitoring frequencies Monitoring frequencies Sampling techniques and protocols Sampling techniques and protocols

32. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 32 Synthesis Scope – Issues (Cont.) Additional features collected Additional features collected Benefits of automation Benefits of automation Equipment specifications Equipment specifications Costs of automated data collection Costs of automated data collection Limitation of technologies Limitation of technologies States’ near-term plans States’ near-term plans

33. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 33 Questionnaire Responses Agencies Responding States – 42 States – 42 District of Columbia District of Columbia FHWA offices – 2 FHWA offices – 2  LTPP  Eastern federal lands Provinces/territories - 10 Provinces/territories - 10 Transport Canada (airfields) Transport Canada (airfields)

34. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 34 Agencies Employing Automated Collection. Cracking, patching – 30 Cracking, patching – 30 IRI - virtually all agencies (HPMS) IRI - virtually all agencies (HPMS) Rut depth – 47 Rut depth – 47 Joint faulting - 23 Joint faulting - 23 Other (ROW, etc.) - 28 Other (ROW, etc.) - 28

35. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 35 Typical Data Collection Vehicle (Courtesy International Cybernetics)

36. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 36 Data Issues – Each Distress/item When collected When collected  Concurrent with other distresses  Separately Sampling interval Sampling interval Location-referencing Location-referencing Methodology of capture Methodology of capture

37. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 37 Data Issues (Cont.) Data management Data management  Platform  Media  Hardware  Quality assurance Data processing Data processing  Automated  Manual Applicable protocols Applicable protocols

38. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 38 Summary of Cracking/patching) Data CollectionMethods Manual collection – 24 Manual collection – 24 Automated collection – 30 Automated collection – 30  Analog  Photographic – 1  Video - 14  Digital (disks & tapes)  Area scan – 17  Line scan - 2

39. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 39 Digital Imaging Technologies (Courtesy Kelvin Wang) Area Scan Line Scan

40. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 40 2048 Pixel (Jpeg) –Area Scan Image (Courtesy Kelvin Wang) 2048 Pixels (transversely), About 3-meters Wide

41. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 41 4096 Pixel (Jpeg) – Line Scan Image (Courtesy Kelvin Wang) 4096 Pixels (transversely), About 4-meters Wide

42. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 42 Line-scan Image With Shadows (Courtesy Virginia DOT)

43. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 43 Summary of Cracking/patching) Data Processing Methods Manual – 20 Manual – 20  Manual data reduction from tapes, etc. Semi-automated – 5 Semi-automated – 5  Significant human intervention Fully automated – 8 Fully automated – 8  Minimum of human intervention

44. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 44 Pavement Surface Distress Analysis System (Courtesy Kelvin Wang)

45. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 45 Example Image and Crack Map (Courtesy Roadware, Inc.) Example Digital Image Example Crack Map

46. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 46 Summary of IRI Data Collection Methods/measurement Techniques Laser - 44 agencies Laser - 44 agencies Acoustic - 3 agencies Acoustic - 3 agencies Infrared - 4 agencies Infrared - 4 agencies Most use AASHTO protocol or variation thereof Most use AASHTO protocol or variation thereof

47. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 47 AASHTO Rut Depth Measurement

48. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 48 Summary of Rut Depth Data Collection Methods 3 sensor – 13 agencies 3 sensor – 13 agencies 5 sensor – 16 agencies 5 sensor – 16 agencies  Generally AASHTO protocol >5 sensors – 13 agencies >5 sensors – 13 agencies  (Usually 30+ sensors)

49. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 49 Scanning Laser (Courtesy Mandli Communications)

50. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 50 AASHTO Joint Faulting Measurement

51. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 51 Summary of Joint Faulting Data Collection Methods Manual – 11 agencies Manual – 11 agencies  Georgia fault meter  Other Sensor - 23 agencies Sensor - 23 agencies (Usually same technology as IRI/rut depth) Few (4 agencies) use AASHTO protocol Few (4 agencies) use AASHTO protocol

52. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 52 Summary of Peripheral Data Collected (No. Agencies) ROW images – 25 ROW images – 25 Sign inventory – 6 Sign inventory – 6 Drainage inventory – 3 Drainage inventory – 3 Other – 11 Other – 11 (Usually shoulders or geometrics)

53. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 53 Contracting Procedures RFP – 18 agencies RFP – 18 agencies RFQ – 7 agencies RFQ – 7 agencies Contract low bidder – 8 agencies Contract low bidder – 8 agencies Typical contract is two years with option to extend Typical contract is two years with option to extend Some contract cracking, do sensor work in- house, some the opposite Some contract cracking, do sensor work in- house, some the opposite Need standardized approach/specifications Need standardized approach/specifications

54. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 54 Quality Assurance QA contract provisions – 22 agencies QA contract provisions – 22 agencies Price adjustment clauses – 12 agencies Price adjustment clauses – 12 agencies QA on sensor data much ahead of cracking data QA on sensor data much ahead of cracking data Need typical variability data Need typical variability data Need better definition of acceptable quality data Need better definition of acceptable quality data

55. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 55 Summary of Costs (Collection and Processing – U. S. $/Mile) Surface distress – $16.00 avg. (3 agencies) Surface distress – $16.00 avg. (3 agencies) IRI – $6.12 avg. (2 agencies) IRI – $6.12 avg. (2 agencies) Rut depth – $1.68 avg. (2 agencies) Rut depth – $1.68 avg. (2 agencies) Joint faulting – $2.23 (1 agency) Joint faulting – $2.23 (1 agency) Combined sensor data – $12.63 avg. (6) Combined sensor data – $12.63 avg. (6) Combined sensor and surface distress – $50.02 avg. (11 agencies) Combined sensor and surface distress – $50.02 avg. (11 agencies)

56. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 56 Benefits of Automation Reduces manpower requirements Reduces manpower requirements Greater safety and more efficiency Greater safety and more efficiency Images provide permanent record Images provide permanent record More consistent data (some see the opposite) More consistent data (some see the opposite) Able to sample 100%, couldn’t before Able to sample 100%, couldn’t before

57. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 57 Research Needs Identified Development of a standard automated data collection approach or “toolbox” Development of a standard automated data collection approach or “toolbox” Study of automated data item collection standards or protocols Study of automated data item collection standards or protocols Study of automated surface distress processing standards or protocols Study of automated surface distress processing standards or protocols Development of quality management programs for collection and processing Development of quality management programs for collection and processing

58. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage C. Bertrand Texas Department of Transportation TxDOT’s Automated Pavement Distress Measurement System

59. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 59 TxDOT’s Automated Pavement Distress Measurement System Designed to provide real-time summary data for PMIS evaluations of network. Designed to provide real-time summary data for PMIS evaluations of network. No human intervention required. No human intervention required. Can be used for project level pavement evaluations. Can be used for project level pavement evaluations.

60. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 60 Measurement Sub-Systems Inertial Profile Inertial Profile Acoustic Rut (moving to scanning laser) Acoustic Rut (moving to scanning laser) Texture for Est. Skid Texture for Est. Skid GPS GPS

61. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 61 Measurement Sub-Systems Digital R-O-W images Digital R-O-W images Distance Distance PMIS Headers with automatic update PMIS Headers with automatic update Mapping Mapping Automated Surface Distress Automated Surface Distress

62. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 62 TxDOT’s Automated Pavement Distress Measurement Vehicle

63. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage Developed by B. Xu and H. Huang University of Texas at Austin Automated Pavement Surface Distress Measurement System

64. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 64 Automated Pavement Distress Measurement System Safety of TxDOT employees and driving public Safety of TxDOT employees and driving public Repeatability and reproducibility Repeatability and reproducibility Data summary density Data summary density Cost Cost

65. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage Automated Surface Distress The system Consists of one CPU computer, one linescan camera, a frame grabber, and custom design software; Consists of one CPU computer, one linescan camera, a frame grabber, and custom design software; Scans a 12 foot lane with 100% coverage at highway speed; Scans a 12 foot lane with 100% coverage at highway speed; Detects sealed and unsealed cracks; Detects sealed and unsealed cracks; Works under various weather conditions. Works under various weather conditions.

66. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage Automated Surface Distress The system Summarizes every 0.1 mile Summarizes every 0.1 mile (ACP) Historic PMIS distress types & (ACP) Historic PMIS distress types & AASHTO protocol AASHTO protocol (Rigid) Jointed and continuous the same & (Rigid) Jointed and continuous the same & AASHTO protocol only AASHTO protocol only

67. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage Image Segmentation Original image Vertical projection Background image Image with high threshold Image with low threshold Subtraction Addition Filtering Sealed crackUnsealed crack

68. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage Longitudinal Crack Detection Original Equalized Traced

69. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage Transverse Crack Detection Original Equalized Traced Shadow

70. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 70 Crack Maps Unsealed CrackSealed Crack

71. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage Performance Tests of Automated Pavement Surface Distress Measurement System

72. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 72 Multiple Scans Longitudinal Cracks on SL360 R1 0 50 100 150 200 250 300 151101151201 Interval (0.1 mile) Feet #1 #2 #3

73. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 73 Multiple Scans Transverse Cracks on SL360 R1 Count 0.0 2.0 4.0 6.0 151101151201 #1 #2 #3 Interval (0.1 mile)

74. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage Correlation of Multiple Scans Longitudinal Crack on SL360 R1 R 2 = 0.9519 0 20 40 60 80 100 120 020406080100120 1 st Scan (feet) 2 nd Scan (feet)

75. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage %CV of Multiple Scans 10.77.2Average 12.37.2 FM3394 420 to 422 14.06.4 FM972 K1 552 to 552+1.5 N/A6.1 FM95 K1 412 to 414 11.79.5 Sl360 L1 436 to 432 4.77.0 Sl360 R1 432 to 436 Transverse Cracks Longitudinal Cracks Pavement Section

76. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 76 Weather Conditions Longitudinal Crack on SL360 R1 0 50 100 150 200 250 300 151101151201 May 8 (heavily cloudy) May 9 (sunny) May 20 (cloudy) Feet Interval (0.1 mile)

77. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 77 Vehicle Speed Longitudinal Crack on SL360 R1 432 to 432+0.5 Feet 0 40 80 120 160 200 050010001500200025003000 Scanned Section (ft) 35 mph 45 mph 55 mph

78. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 78 Mopac & SL360 Loop Scan 0 50 100 150 200 0.01.02.03.04.05.06.07.0 8.0 longitudinal crack 0 50 100 150 200 8.09.010.011.012.013.014.015.0 16.0 Miles 0 50 100 150 200 16.017.018.019.020.021.022.023.024.0

79. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 79 Multiple Scans of 3288 Transverse crack Interval (0.1 mile)

80. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 80 Multiple Scans of 3288 longitudinal crack Interval (0.1 mile)

81. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 81 Correlation Coefficients of Multiple Runs Vehicles Runs 1005732833288 translongtranslongtranslong R1 & R20.92080.75310.92880.88750.95940.7567 R1 & R30.90370.64860.91130.86820.86590.7681 R2 & R30.90620.63430.91220.88440.92040.821

82. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 82 Scans of Multiple Vehicles Transverse crack Interval (0.1 mile)

83. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 83 Scans of Multiple Vehicles Longitudinal crack Interval (0.1 mile)

84. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 84 Correlation Coefficients of Multiple Vehicles TransLong 10057 to 32830.73960.6678 10057 to 32880.91990.9037 3283 to 32880.73420.684

85. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 85 Long Distance Scans Transverse crack Interval (0.1 mile)

86. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 86 Long Distance Scans Longitudinal crack Interval (0.1 mile)

87. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 87 Correlation Coefficients of Long Distance Scans TransLong 10057 to 32830.95110.9784 10057 to 32880.95270.9837 3283 to 32880.89910.9691

88. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 88 Conclusions Repeatability: R 2 >0.8 (long distance), CV 5.0~14% Repeatability: R 2 >0.8 (long distance), CV 5.0~14% 100% distance coverage 100% distance coverage 10-12 feet lane coverage 10-12 feet lane coverage Real-time image processing Real-time image processing 5-70 mph of vehicle speed 5-70 mph of vehicle speed Simple setup (one CPU & one camera) Simple setup (one CPU & one camera) Weather tolerance (cloudy or sunny) Weather tolerance (cloudy or sunny) PMIS and AASHTO protocols PMIS and AASHTO protocols

89. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 89 Next Phase of Implementation Conversion of the remaining fleet (13) Conversion of the remaining fleet (13) Conversion of PMIS pavement condition scores Conversion of PMIS pavement condition scores Development of lighting system Development of lighting system Identification of additional distress features Identification of additional distress features Production of crack maps at highway speeds Production of crack maps at highway speeds

90. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage Manual Image Analysis System (MIAS) as QA/QC Tool for Automated Image Analysis System (AIAS) Workshop 134 Sunday, January 11, 2004, 2:00 PM - 5:30 PM, Marriott Sponsored by: A2B06 - Pavement Monitoring, Evaluation and Data Storage Hosin “David” Lee, Ph.D., P.E. Public Policy Center Civil and Environmental Engineering University of Iowa Email: hlee@engr.uiowa.edu

91. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 91 QA/QC Process Use Visual Condition Survey as Ground-truth Visual Condition Survey Automated Image Analysis System Pavement Distress Information QA/QC Pavement Management System Manual Image Analysis System

92. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 92 Automatic Image Collection System #1 Courtesy of Waylink System, Inc.

93. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 93 Pavement Image from AICS #1 Resolution: 4096  2048 pixels Resolution: 4096  2048 pixels Courtesy of WayLink System, Inc.

94. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 94 Automatic Image Collection System #2 Courtesy of Roadware Inc.

95. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 95 Pavement Image from AICS #2 Courtesy: Roadware Inc. Resolution: Resolution: 1232  3180 pixels 1232  3180 pixels

96. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 96 Automatic Image Collection System #3-1 Courtesy: Samsung SDSA

97. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 97 Automatic Image Collection System #3-2

98. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 98 Automatic Image Collection System #3-3 Courtesy: APSA

99. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 99 Pavement Image from AICS #3 Courtesy: Samsung SDSA Resolution: 776 x 582 pixels Resolution: 776 x 582 pixels

100. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 100 Visual Condition Survey As Ground-Truth?

101. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 101 Automated Image Analysis System #1  Parallel Processing at Real Time Courtesy: WayLink System, Inc.

102. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 102 AIAS #1 Input Parameters  Automatically tuned  Thresholdings  Breakdowns of area sizes  Levels of connectivity

103. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 103 AIAS # 1 vs. Visual Condition Survey Courtesy: WayLink Systems Co. Visual, 5% of each mile Automated, 100% of each mile Standard Deviation

104. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 104 Automated Image Analysis System #2  Generation of Crack Maps Courtesy: Roadware Group Inc.

105. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 105 AIAS #2 Input Parameters Courtesy: Roadware Group Inc.

106. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 106 Crack Length Measured by MIAS and AIAS # 2 Raw Score MIASFull AIAS#2Semi AIAS#2 Image #(Ground Truth)(No manual intervention)(Noise Removed) (Meter) 01.jpg 28.1327.0224.11 02.jpg 19.7617.99 03.jpg 59.5452.0451.36 04.jpg 18.8118.10 05.jpg 29.4327.8327.50 06.jpg 19.7119.50 07.jpg 77.7672.02 08.jpg 29.9730.1528.46 09.jpg 26.7026.7526.13 10.jpg 97.3589.8889.36 Courtesy: Roadware Group Inc.

107. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 107 AIAS # 2 (semi-automated) vs. MIAS Comparison Results Image No. Semi vs. Full Automation (m) (%) Manual vs. Semi-Auto (m) (%) 01.jpg -2.9010.7%-4.02-14.3% 02.jpg 0.000.0%-1.76-8.9% 03.jpg -0.681.3%-8.18-13.7% 04.jpg 0.000.0%-0.72-3.8% 05.jpg -0.331.2%-1.92-6.5% 06.jpg 0.000.0%-0.20-1.0% 07.jpg 0.000.0%-5.74-7.4% 08.jpg -1.695.6%-1.51-5.0% 09.jpg -0.622.3%-0.57-2.1% 10.jpg -0.520.6%-7.99-8.2% Courtesy: Roadware Group Inc.

108. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage Fatigue Crack Length and Severity Automated Image Analysis System #3 Courtesy: Samsung SDSA

109. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 109 AIAS #3 Input Parameters Courtesy: Samsung SDSA

110. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 110 Manual Image Analysis System #3 Starting point of lon gitudinal cracking Ending point Courtesy: Samsung SDSA

111. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 111 MIAS #3 for Measuring Crack Areas Measure the crack area Courtesy: Samsung SDSA

112. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 112 MIAS #3 for Measuring Crack Width Measure crack width for severity Courtesy: Samsung SDSA

113. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage Crack Length Crack Type Crack Severity MIAS #3 Output Courtesy: Samsung SDSA

114. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 114 Crack Length and Width Measured by MIAS #3 Image Name Manual Analysis (Image-Based Ground Truth) Inside Wheel Path Crack Between Wheel Path Crack Outside Wheel Path Crack Intensity (m/m 2 ) Width (mm) Intensity (m/m 2 ) Width (mm) Intensity (m/m 2 ) Width (mm) 001.jpg0.6211.00.598.10.548.5 002.jpg6.3513.26.0810.86.2113.2 003.jpg0.557.40.5613.20.608.4 004.jpg1.8616.41.5920.80.6211.2 005.jpg0.3610.00.838.50.559.6 006.jpg0.517.31.237.20.9910.7 007.jpg1.9722.30.517.40.000 008.jpg0.5410.80.5712.30.5518.1 009.jpg0.6116.42.01111.868.1 010.jpg2.3720.80.5516.40.000 Average1.5713.561.4511.571.198.78 Courtesy: Samsung SDSA

115. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 115 Crack Length and Width Measured by AIAS # 3 Image Name Automatic Analysis (No manual intervention) Inside Wheel Path Crack Between Wheel Path Crack Outside Wheel Path Crack Intensity (m/m 2 ) Width (mm) Intensity (m/m 2 ) Width (mm) Intensity (m/m 2 ) Width (mm) 001.jpg0.77100.4111.80.717.9 002.jpg6.6911.55.9813.16.1011.8 003.jpg0.6515.20.41190.6515.9 004.jpg2.1914.10.7718.41.1214 005.jpg0.367.90.9514.80.5314.6 006.jpg0.599.90.9513.51.4814.5 007.jpg2.7218.10.415.90.000 008.jpg0.4113.70.7111.80.5912.7 009.jpg0.5914.51.2410.31.898.9 010.jpg2.4322.70.3017.20.000 Average1.7413.761.2113.581.3110.03 Courtesy: Samsung SDSA

116. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 116 AIAS #3 vs. MIAS #3 (Average Bias) Image Name Comparison (Manual vs. Auto) Inside Wheel Path Crack Between Wheel Path Crack Outside Wheel Path Crack Intensity (m/m 2 ) Width (mm) Intensity (m/m 2 ) Width (mm) Intensity (m/m 2 ) Width (mm) 001.jpg-0.151.00.18-3.7-0.170.6 002.jpg-0.341.70.10-2.30.111.4 003.jpg-0.10-7.80.15-5.8-0.05-7.5 004.jpg-0.332.30.822.4-0.50-2.8 005.jpg0.002.1-0.12-6.30.02-5.0 006.jpg-0.08-2.60.28-6.3-0.49-3.8 007.jpg-0.754.20.101.50.000.0 008.jpg0.13-2.9-0.140.5-0.045.4 009.jpg0.021.90.770.7-0.03-0.8 010.jpg-0.06-1.90.25-0.80.000.0 Average-0.17-0.200.24-2.01-0.12-1.25 Courtesy: Samsung SDSA

117. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 117 Evaluation of AIAS Systems Own AICS? AIAS #1 AIAS #2 IBI Precision? IBI Bias? MIAS Available? Yes N/A N/A N/A 4096x2048 Yes 7.1% -7.1% Yes 1232X3180 AIAS #3 No 19.6% 10.8% Yes 776X582 AICS Resolution?

118. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 118 Summary and Conclusion Adoption of MIAS/AIAS As Standard Manual Visual Condition Survey Not as Ground-truth Manual Image Analysis System (MIAS) as Ground-truth MIAS as QA/QC tool for Automated Image Analysis System (AIAS) Image by Image Precision and Bias Analysis AIAS for any Automated Image Collection System 1 2 3 4 5

119. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 119 University of Iowa Public Policy Center

120. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 120 QC/QA Processes for LTPP Photographic Distress Data Gonzalo Rada and Amy Simpson MACTEC Engr. & Consulting, Inc. John Hunt ERES Consultants

121. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 121 What is the LTPP Program? 20-year study whose goal is to provide data to explain “how and why pavements perform as they do” 20-year study whose goal is to provide data to explain “how and why pavements perform as they do” 2,500 test sections throughout North America 2,500 test sections throughout North America Data collected at each test section includes inventory, materials, performance monitoring, traffic, climatic, M&R Data collected at each test section includes inventory, materials, performance monitoring, traffic, climatic, M&R

122. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 122 Test Section Locations

123. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 123 Distress Data: Purpose Detailed, distress specific condition data for use in development/ validation of performance models Detailed, distress specific condition data for use in development/ validation of performance models Permanent, objective, high- resolution record of pavement condition Permanent, objective, high- resolution record of pavement condition

124. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 124 Distress Data: Collection

125. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 125 Distress Data: Collection

126. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 126 1989 - 1996  5 rounds providing section coverage on ~2-year cycle 1989 - 1996  5 rounds providing section coverage on ~2-year cycle 1996 - 1999  only manual distress surveys due to funding 1996 - 1999  only manual distress surveys due to funding 1999 - date  4 rounds providing coverage of 1,000 sections/year 1999 - date  4 rounds providing coverage of 1,000 sections/year 9 rounds completed to date; Round 10 ~80% complete Distress Data: Collection

127. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 127 Distress Data: Interpretation vement Distress Analysis System Pavement Distress Analysis System

128. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 128 Distress Data: Interpretation

129. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 129 Distress Data: Interpretation

130. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 130 Distress Data: Interpretation Asphaltic Concrete

131. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 131 Accurate, reliable distress data vital to success of LTPP Accurate, reliable distress data vital to success of LTPP QC/QA processes QC/QA processes  Prior to data collection  During data collection  During data interpretation  After data interpretation  Others Distress Data: QC/QA

132. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 132 Prior to Data Collection

133. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 133 Prior to Data Collection Minimum attendance requirements Written & film interpretation examination

134. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 134 Additional requirements Distress raters must perform 36 interpretations/year Distress raters must perform 36 interpretations/year QC reviewers must perform 24 interpretations/year QC reviewers must perform 24 interpretations/year Raters and QC reviewers must be accredited annually Raters and QC reviewers must be accredited annually Prior to Data Collection

135. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 135 During Data Collection Equipment calibration (before each trip) DMI calibration DMI calibration Surface distress system calibration Surface distress system calibration  Illumination  Longitudinal distortion Transverse profile system (block and hairline placement) Transverse profile system (block and hairline placement)  Resolution

136. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 136 During Data Collection

137. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 137 QC checks (weekly/every 20 sections) Actual versus planned surveys Actual versus planned surveys Quality of surface distress images Quality of surface distress images  Longitudinal distortion  Lateral lane displacement  Film exposure  Resolution board checks Quality of transverse profile film Quality of transverse profile film  Block calibration  Skipped frames  Transverse profile placement During Data Collection

138. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 138 FC = film, TC = technical support, and RC = regional contractors During Data Interpretation

139. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 139 QC process QC process  Sections rated  Comparison plots generated  100% QC review  Changes noted and passed to QA reviewer QA process QA process  1 in 10 sections randomly selected for review During Data Interpretation

140. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 140 Time-series check Time-series check  DVA software  Subset of distress data  Total quantity only Database checks Database checks  Range checks  Intra-modular and intra-field checks After Data Interpretation

141. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 141 After Data Interpretation

142. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 142 Other QC/QA Elements Annual QA review of regional contractors Annual QA review of regional contractors Data studies Data studies  1999 variability study  2001 data consolidation study Feedback process Feedback process  Operational problem reports  Analysis/operations feedback reports

143. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 143 Summary & Conclusions Accurate, reliable distress data vital to success of LTPP Accurate, reliable distress data vital to success of LTPP Numerous QC/QA processes in place Numerous QC/QA processes in place  Advance planning (past experience)  Lessons learned Interpretation process subjective, but…film provides permanent record Interpretation process subjective, but…film provides permanent record

144. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage Automated Distress Data Collection Workshop 134 Sunday, January 11, 2004, 2:00 PM - 5:30 PM, Marriott Sponsored by: A2B06 - Pavement Monitoring, Evaluation and Data Storage

145. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 145 Multi-Year Automated Pavement Condition Survey in Iowa Omar Smadi CTRE/Iowa State University

146. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 146 Outline Introduction and Background Introduction and Background Automated Distress Data Automated Distress Data Data Collection Plan Data Collection Plan The Iowa Pavement Management Program (IPMP): The Iowa Pavement Management Program (IPMP):  Participation  Sample Data Discussion and Conclusions Discussion and Conclusions

147. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 147 Introduction ISTEA of 1991 ISTEA of 1991 Iowa DOT, cities, and counties in Iowa started the PMS development in 1994 for FAE routes Iowa DOT, cities, and counties in Iowa started the PMS development in 1994 for FAE routes Automated distress equipment evaluation (1995) Automated distress equipment evaluation (1995) Automated distress data collection (1996) Automated distress data collection (1996)

148. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 148 Equipment Evaluation Different Vendors were invited Different Vendors were invited Discussion of the technology used: Discussion of the technology used:  For data collection  For data processing Quality of the data (control sites) Quality of the data (control sites) Experience with state DOT contracts Experience with state DOT contracts Cost Cost

149. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 149 Data Quality 8 control sites were selected: 8 control sites were selected:  0.50 km long  4 ACC (full depth and composite) surface and 4 PCC surface  State, city, and county segments Manual survey conducted: Manual survey conducted:  Detailed crack and patch data IRI and Rutting using DOT equipment IRI and Rutting using DOT equipment

150. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 150 Automated Distress Data (PCC) D-Cracking:Number of Joints SHRP Moderate and High D-Cracking:Number of Joints SHRP Moderate and High Joint Spalling:Number of Joints SHRP Moderate and High Joint Spalling:Number of Joints SHRP Moderate and High T-Cracking:Number SHRP Low, Moderate and High T-Cracking:Number SHRP Low, Moderate and High Patching:Area and Number Distress or No-Distress Patching:Area and Number Distress or No-Distress Ride:IRI Ride:IRI

151. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 151 Automated Distress Data (ACC) T-Cracking:Number SHRP Low, Moderate & High T-Cracking:Number SHRP Low, Moderate & High L-Cracking:Length SHRP Low, Moderate & High L-Cracking:Length SHRP Low, Moderate & High Block & Alligator:Area SHRP Moderate and High Block & Alligator:Area SHRP Moderate and High Potholes:Number Potholes:Number Patching:Area and Number Distress or No-Distress Patching:Area and Number Distress or No-Distress Ride and Rutting Ride and Rutting

152. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 152 Data Collection Plan 2-year cycle for data collection 2-year cycle for data collection  Started with 12,000 miles  Currently over 30,000 miles Data quality QC/QA Process Data quality QC/QA Process  Control sites  Random sites  HPMS segments 10 meter test segment 10 meter test segment

153. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 153 Data Collection Plan (Cont.) Routes to be collected: Routes to be collected:  Literal description  GIS Vendor delivers: Vendor delivers:  GPS coordinates (LAT and LONG)  Route Name  Jurisdiction  Distress data based on 10 m

154. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 154 QC/QA Process 8 control sites: 8 control sites:  Manual data at beginning of project  Automated distress data at beginning and monthly until the end of the project Random Sites: Random Sites:  Manual data collected  Comparison with what vendor delivered  Has been completed twice since 1996

155. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 155 QC/QA Process (Cont.) Iowa DOT started this process in 2001 Iowa DOT started this process in 2001 200 Random sections are selected 200 Random sections are selected  Pavement images are purchased  Distress data collected  Comparison with what vendor provided

156. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 156 QC/QA Process (Cont.) Location of test segments (GPS) Location of test segments (GPS) Coverage of the system Coverage of the system Proper jurisdiction Proper jurisdiction Proper surface type Proper surface type

157. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 157 Iowa Pavement Management Program (IPMP) Support of the MANAGEMENT, PLANNING, and PROGRAMMING needs of transportation agencies Support of the MANAGEMENT, PLANNING, and PROGRAMMING needs of transportation agencies Provide pavement management information, tools, and training supporting both PROJECT level and NETWORK level pavement management activities Provide pavement management information, tools, and training supporting both PROJECT level and NETWORK level pavement management activities

158. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 158 Data Collection 1996-1997 Participation 2 7 1 8 119 5 12 16 10 6 15 17 14 13 4 3 18 Sioux City Des Moines Council Bluffs Davenport Dubuque Cedar Rapids Iowa City Waterloo Yes (1996) Yes (1997)

159. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 159 Data Collection 1996-1999 Participation 2 7 1 8 11 9 5 12 16 10 6 15 17 14 13 4 3 18 Sioux City Des Moines Council Bluffs Davenport Dubuque Cedar Rapids Iowa City Waterloo Yes (96 & 97) Yes (98 & 99)

160. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 160 Data Collection 1996-2003 Participation 2 7 1 8 11 9 5 12 16 10 6 15 17 14 13 4 3 18 Sioux City Des Moines Council Bluffs Davenport Dubuque Cedar Rapids Iowa City Waterloo Yes (96 & 97) Yes (98 & 99) 23 Yes (2000)

161. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 161 Mileage Statistics (Statewide)

162. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 162 Municipal Testing (Non-FAE) 1999 2000 23 2001-2003

163. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 163 IPMP Sample Data Atlantic, Iowa

164. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 164 IPMP GIS Tools

165. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 165 IPMP Additional Products

166. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 166 IPMP Additional Products

167. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 167 PMS Software

168. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 168 PMS Software

169. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 169 PMS Software

170. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 170 PMS Software

171. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 171 PMS Software

172. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 172 Conclusion Automated distress data is feasible Automated distress data is feasible Completed 4 cycles of data collection (a total of 90,000 miles) Completed 4 cycles of data collection (a total of 90,000 miles) Additional agencies are participating in the system Additional agencies are participating in the system One component in the PMS puzzle One component in the PMS puzzle

173. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 173 Automated Crack Detection in the State of Maryland Jonathan L. Groeger Axiom Decision Systems, Inc.

174. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 174 Agenda Overview Overview Development Process Development Process Data Collection Procedures Data Collection Procedures Quality Assurance Quality Assurance Results Results

175. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 175 Maryland Pavement Network 16,000 lane miles 16,000 lane miles $100 million pavement preservation budget $100 million pavement preservation budget Seven districts Seven districts New preservation strategy more objective New preservation strategy more objective Performance-based criteria Performance-based criteria

176. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 176 Problem Statement Cracking data not collected recently Cracking data not collected recently Data needed for PMS performance modeling Data needed for PMS performance modeling Very limited resources Very limited resources Existing technology not proven at Maryland Existing technology not proven at Maryland Quality is #1 Quality is #1

177. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 177 Resources ARAN data collection vehicle ARAN data collection vehicle WiseCrax crack detection software WiseCrax crack detection software AASHTO Cracking Protocol AASHTO Cracking Protocol Pavement Management Division staff Pavement Management Division staff Consultant resources Consultant resources

178. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage Development Process

179. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 179 Process Affirmation from Connecticut DOT Affirmation from Connecticut DOT Pilot Study Pilot Study Benchmark Survey Benchmark Survey Production Testing Production Testing

180. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 180 Pilot Study Gained experience with Wisecrax Gained experience with Wisecrax Gained experience with AASHTO cracking protocol Gained experience with AASHTO cracking protocol Determined tentative condition rating scheme Determined tentative condition rating scheme Hardware problems Hardware problems More work to be done!!! More work to be done!!!

181. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 181 Benchmark Survey ARAN is viable data collection platform ARAN is viable data collection platform WX is viable processing tool WX is viable processing tool Manufacturer versus MD SHA results similar Manufacturer versus MD SHA results similar AASHTO protocol “with a twist” chosen as data processing method AASHTO protocol “with a twist” chosen as data processing method Field versus automated comparison very encouraging Field versus automated comparison very encouraging

182. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 182 Production Testing Process Process  Perform crack survey for one district  Submit to district personnel for validation Results Results  Validated processes  Data deemed reasonable  Ready to Roll!

183. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage Data Collection Procedures

184. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 184 Data Collection Equipment ARAN vehicle ARAN vehicle High-resolution digital video High-resolution digital video Ultrasonic sensors (rutting) Ultrasonic sensors (rutting) Laser sensors (profile) Laser sensors (profile) Accelerometer Accelerometer Gyroscope Gyroscope Global positioning system Global positioning system Distance measuring device Distance measuring device

185. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 185 ARAN 10,000 lane miles State equipment/personnel 6 month period +/- Data Collection

186. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage Data Collection Elements Digital ROW Video Digital Pvmt Images Rutting Grade Asset Location Ride (IRI) Cross-slope Curve Radius

187. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 187 7 Step Process Pre-processing Processing Quality Control Quality Assurance Classification/ Rating Data Reduction Data Management Data provided by field crew weekly Data provided by field crew weekly ARAN log sheets entered in database ARAN log sheets entered in database Data cross-checked with PMS records Data cross-checked with PMS records Removable hard drives downloaded and archived Removable hard drives downloaded and archived Field progress determined Field progress determined Office progress tracked Office progress tracked Step 1 Step 2 Step 3 Step 4 Step 5 Step 6 Step 7

188. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 188 Step 2. Pre-processing Data loaded into Wisecrax computer from archive tapes Data loaded into Wisecrax computer from archive tapes - Control file - JPG images Wisecrax initiated Wisecrax initiated Images examined Images examined - Image present - Lighting even - Clear image Detection parameters set Detection parameters set New overlays (< 2 years) ignored New overlays (< 2 years) ignored

189. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 189 13 – 17 mph Automated 30-60 mile batches Step 3. Processing (detection)

190. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 190 Completeness Quality (> 80% crack detection) Trends Close to 100% check Step 4. Quality Control

191. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 191 Weekly 10 percent sample Check of overall process Sampling Approach Step 5. Quality Assurance

192. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 192 Long/Trans, Low, Med, High Fully Automated, 800 mph AASHTO Protocol Step 6. Classify/Rate

193. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 193 Step 7. Data Reduction Large amount of data generated Large amount of data generated 1000 or more text files 1000 or more text files 1,000,000 lines of data (records) 1,000,000 lines of data (records) Resultant data is in raw cracking form Resultant data is in raw cracking form Neither intuitive or easy to use in PMS Neither intuitive or easy to use in PMS

194. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 194 Data Reduction QC/QA Summarize to 0.1 mile Cracking Data Output to PMS Progress Reports Assign Condition State Data Completeness Data Completeness Range Checks Range Checks Logic Checks Logic Checks Trend Analysis Trend Analysis

195. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 195 Maryland Process Developed procedure to convert raw cracking data to condition index Developed procedure to convert raw cracking data to condition index 0 to 100 index 0 to 100 index Very Good >= 90 Very Good >= 90 Good 80-89 Good 80-89 Fair 65-79 Fair 65-79 Mediocre50-64 Mediocre50-64 Poor<50 Poor<50 Use AASHTO cracking breakpoints and PCI deduct method Use AASHTO cracking breakpoints and PCI deduct method Recently re-verified condition breakpoints with field observations Recently re-verified condition breakpoints with field observations

196. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage Quality Assurance

197. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 197 Quality Assurance Data Completeness Data Completeness Range Checks Range Checks Data Reasonableness Data Reasonableness Trends Trends Field Verification Field Verification

198. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 198 Quality Assurance

199. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 199 Quality Assurance Fail Pass

200. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage Results

201. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 201 Consistency

202. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 202 Overall Results

203. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 203 Reporting Network Condition Rutting Friction Ride Cracking Overall Condition Index

204. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 204 Performance Modeling 0% 20% 40% 60% 80% 100% Forecasting Condition Performance Models Benefit Calculation M & R Strategy Corrective Maintenance Thin Overlay Preventive Maintenance Thick Overlay RSL Remaining Service Life

205. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage Lessons Learned Keep it Simple Validate Your Data Viable Technology Partner with Data Collection Vendor Need Commitment from Above Large Resource Commitment Rigorous QC/QA a Must Phased Approach is the Key

206. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 206 Future Directions of Technology Development Kelvin C.P. Wang University o f Arkansas

207. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 207 Protocols and New Technologies Cracking Protocols Cracking Protocols  Universal Cracking Indicator (CI)  AASHTO Interim Distress Protocol New Technologies New Technologies  Laser Based 3D Scanning  Stereovision

208. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 208 Use of Universal Cracking Indicator, UCI William D. Paterson of World Bank Devised A New Method to Calculate A Distress Index William D. Paterson of World Bank Devised A New Method to Calculate A Distress Index  Objective and Universally Transferable  Able to Be Applied for Differing Pavement Types, Environments, and Standards  Measurable by a Variety of Methods at Differing Levels of Detail and Precision  Progressive, not Constrained by Present Technology  Simple

209. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 209 Five Attributes for Cracking in UCI Extent: Area Covered by Cracking, Unit: Area, or % of Total Area Extent: Area Covered by Cracking, Unit: Area, or % of Total Area Severity: Average Width of Crack, Unit: Level or Millimeters Severity: Average Width of Crack, Unit: Level or Millimeters Intensity: Length of Cracks per Unit Area Intensity: Length of Cracks per Unit Area Pattern or Type: Orientation & Interconnectedness of Cracks (Alligator, Block, etc…) Pattern or Type: Orientation & Interconnectedness of Cracks (Alligator, Block, etc…) Location: Wheel Path, Edge, Joint, or Random Location: Wheel Path, Edge, Joint, or Random

210. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 210 Cracking Indicator, CI CI: Dimensionless CI: Dimensionless Purely A Physical Dimension Measure Purely A Physical Dimension Measure No Prior Dependency on Cause or Type of Cracking No Prior Dependency on Cause or Type of Cracking Pattern & Locations Parameters are Useful Pattern & Locations Parameters are Useful  In Diagnosing Causes  In Determining Maintenance Needs and Quantities of Maintenance and Repair CI = Extend x Intensity (m/m 2 ) x Crack Width (mm)

211. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 211 Simple Illustration of Using CI A Pavement with CI of 3,200 May Consists of 2,000 Alligator, 700 Longitudinal, and 500 Irregular Cracking A Pavement with CI of 3,200 May Consists of 2,000 Alligator, 700 Longitudinal, and 500 Irregular Cracking  The Numbers are Additive, Even When Areas are Overlapping For Location, A CI of 3,200 May Include 2,300 in Outer Wheel Path, and 900 not in Wheel Paths For Location, A CI of 3,200 May Include 2,300 in Outer Wheel Path, and 900 not in Wheel Paths

212. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 212 Simple, Subdivisibility, Additivity A Total CI of 3,200 for A Pavement Section A Total CI of 3,200 for A Pavement Section  2,000 Alligator in Outer Wheel Path  300 Irregular in Outer Wheel Path  200 Irregular not in Wheel Path  200 Longitudinal on Centerline  500 Longitudinal at Edge Application of CI Application of CI  Not As Universal As PSI Yet

213. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 213 Computational Example for Longitudinal CI Longitudinal Crack (l L, w L ) Transverse Crack (l T, w T ) Alligator Cracking (l A, w A ) a c b l=Length, w=Width, Total Areas, A=a+b

214. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 214 Computational Example for Longitudinal CI Longitudinal Crack (l L, w L ) Transverse Crack (l T, w T ) Alligator Cracking (l A, w A ) a c b l=Length, w=Width, Total Areas, A=a+b

215. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 215 Computational Example for Alligator CI Longitudinal Crack (l L, w L ) Transverse Crack (l T, w T ) Alligator Cracking (l A, w A ) a c b l=Length, w=Width, Total Areas, A=a+b

216. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 216 Computational Example for Transverse CI Longitudinal Crack (l L, w L ) Transverse Crack (l T, w T ) Alligator Cracking (l A, w A ) a c b l=Length, w=Width, Total Areas, A=a+b

217. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 217 Computational Example for Aggregated CI Longitudinal Crack (l L, w L ) Transverse Crack (l T, w T ) Alligator Cracking (l A, w A ) a c b l=Length, w=Width, Total Areas, A=a+b

218. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 218 Goal of AASHTO PP44-00 Quantifying cracking in asphalt pavement surfaces both in wheel and non wheel path areas Quantifying cracking in asphalt pavement surfaces both in wheel and non wheel path areas Standardization: consistent pavement condition estimates for network pavement management Standardization: consistent pavement condition estimates for network pavement management

219. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 219 Automated and Manual Surveys Automated surveys - Use a vehicle traveling at near highway speeds and collect data on the entire length of roadway (100% sample) Automated surveys - Use a vehicle traveling at near highway speeds and collect data on the entire length of roadway (100% sample) Manual surveys - Observe distresses and record data on a minimum 10% sample. Rating continuous film or tape in an office setting is considered a manual survey Manual surveys - Observe distresses and record data on a minimum 10% sample. Rating continuous film or tape in an office setting is considered a manual survey

220. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 220 Minimum Survey Area Survey a 2.500 m (8 ft) strip in the outside lane as shown (Optional 3.6m (12 ft) full lane width). Survey a 2.500 m (8 ft) strip in the outside lane as shown (Optional 3.6m (12 ft) full lane width). For undivided highways survey one direction For undivided highways survey one direction For each survey cycle, use the same direction(s) of travel and survey lane(s). For each survey cycle, use the same direction(s) of travel and survey lane(s).

221. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 221 Minimum Cracking Definition and Estimation A crack: a discontinuity in the pavement surface, with minimum dimensions of 3 mm (1/8 in) width and 25 mm (1 in) length A crack: a discontinuity in the pavement surface, with minimum dimensions of 3 mm (1/8 in) width and 25 mm (1 in) length  Including longitudinal cracks, transverse cracks, interconnected cracks  Quantifying & differentiating between load associated (fatigue) and non-load associated (environmental, reflective, etc.) pavement cracking and joints.

222. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 222 Wheel-Path Designations

223. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 223 Limitations Sealed cracks: not quantified by manual surveys Sealed cracks: not quantified by manual surveys Automated survey equipment: not to quantify any discontinuity greater than 25 mm (1 in) width Automated survey equipment: not to quantify any discontinuity greater than 25 mm (1 in) width

224. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 224 Classify cracking by severity and intensity Severity Level 1: Cracks < 3mm (1/8 in) Severity Level 2: Cracks > 3mm (1.8 in) and <6 (1/4 in) mm Severity Level 3: Cracks > 6 (1/4 in) mm width. Intensity of each cracking Level: the total length of cracking per unit area (m/m 2 ) for each defined survey strip

225. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 225 Recording of Data Automated surveys - 100% coverage Automated surveys - 100% coverage  The data summary interval: 0.1 km (0.062 mi) Manual surveys - minimum 10% of the section length Manual surveys - minimum 10% of the section length

226. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 226 Quality Assurance Quality Assurance Plan Quality Assurance Plan Qualification and Training Qualification and Training Equipment Equipment Validation Sections Validation Sections Additional Checks Additional Checks

227. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 227 The Shadow Moire Method Pavement Surface h1h1 H n=1 n=2 d CameraLight Source h Plane of Grating Contour Plane h 1 Contour Plane h 2 h2h2 p ∆

228. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 228 Rotation Laser Scanning at 10,000 RPM, Phoenix Scientific

229. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 229 BIRIS Laser Technology, GIE Tech

230. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 230 Swiss CREHOS System

231. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 231 Basis of Stereovision for Condition Survey 4-Meter Wide Pavement Two Cameras, Resolution 4096- Pixel/Camera 2D Crack Detection Overlapping Crack Recognition Establish 3D Surface Model Condition Survey with Geometric Modeling Pavement Surface Condition Database

232. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 232 Long Term Goal Survey Automation for Most of Distress Types Identified Survey Automation for Most of Distress Types Identified All Hardware Items: Commercially Available All Hardware Items: Commercially Available Data Collection and Interpretation at Highway Speed Data Collection and Interpretation at Highway Speed 1-mm Resolution in X, Y, and Z Directions on Entire Pavement Surface 1-mm Resolution in X, Y, and Z Directions on Entire Pavement Surface Providing Geometries of Surface Defects for Inclusion into Proper Indices Providing Geometries of Surface Defects for Inclusion into Proper Indices

233. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 233 Research Conducted in the IDEA Project Setup Dual Camera System in the Existing Vehicle Setup Dual Camera System in the Existing Vehicle Developing Mathematical Models for Stereovision Based on Pairs of Images Developing Mathematical Models for Stereovision Based on Pairs of Images Corrections of Lens Distortion Corrections of Lens Distortion Extracting 3D Surface from Pavement Extracting 3D Surface from Pavement 3D Surface Reconstruction 3D Surface Reconstruction

234. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 234 Hardware Setup

235. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 235 Camera Calibration for Left and Right Images

236. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 236 Calibration Software for Several Coordinate Systems

237. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 237 Ccorrection of Lens Distortion

238. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 238 Image Matching

239. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 239 Matching Example

240. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 240 Reconstruction of 3D Surface

241. Automated Distress Data Collection Workshop, January 11, 2004, TRB, A2B06 - Pavement Monitoring, Evaluation & Data Storage 241 Conclusion Progress Toward Digital Progress Toward Digital Full Automation for Cracking Survey Full Automation for Cracking Survey  Implementation and Validation Quality Control Quality Control  Critical for Successful Application Newer Technologies Newer Technologies  Difficult Problem  Automation of Comprehensive Data Collection and Analysis