Automated collection of AusRAP road attributes using DVR and pattern recognition techniques: Y2 2018/19

This interim report presents the proposed methods, experiments, results and future directions for the project entitled automated collection of AusRAP attributes using DVR, MLS and pattern recognition. The proposed 3-D segmentation and classification method using MLS data and 2-D segmentation and classification method using DVR data for identifying AusRAP attributes are presented. The distance calculation techniques for both MLS and DVR data are described. The proposed methods are implemented in Python programming language and incorporated in development of software for automatically identifying AuSRAP attributes. The proposed methods are tested on a large training and testing data. The experimental results and future directions are presented in this report.

• Record URL:
  https://www.nacoe.com.au/wp-content/uploads/2021/03/R54-ANNUAL-SUMMARY-REPORT-YEAR-2.pdf
• Corporate Authors:

  National Asset Centre of Excellence (NACOE)

  Brisbane, Queensland   
• Authors:
  • Verma, B
  • Affum, J
  • Zhong, M
• Publication Date: 2019-9

Media Info

• Pagination: 19p

Subject/Index Terms

• TRT Terms: Databases; Environment; Highway engineering; Image processing; Pavement maintenance; Pavements; Risk assessment; Tests for suitability, service and quality
• Geographic Terms: Queensland
• ATRI Terms: Database; Image processing; Pavement evaluation; Pavement maintenance; Risk assessment; Road engineering; Road environment
• Subject Areas: Highways; Pavements;

Filing Info

• Accession Number: 01781223
• Record Type: Publication
• Source Agency: ARRB Group Limited
• Report/Paper Numbers: R54
• Files: ATRI
• Created Date: Sep 2 2021 2:25PM