Terrestrial Mobile LiDAR for Roadway Feature Extraction
Terrestrial Mobile LiDAR (TML) is a powerful mapping method that uses laser scanning to identify roadway assets and related infrastructure quickly, accurately and cost efficiently. Using Leica Pegasus:Two Ultimate mobile mapping platform, detailed inventory mobile LiDAR data has been collected spread over 600 centerline miles of interstate highway in Virginia. Current feature extraction workflow was designed to extract on 30+ asset types including signs, guardrails, light poles, overhead utilities and so on. As an important first step in streamlining the process, an comprehensive evaluation of current data and workflow is proposed. The intent is to provide guidelines to help insure efficient use of TML technology and tools in support of road feature extraction, data management and visualization for the LiDAR inventory data. Meanwhile, it is also expected to increase the usability of the data and maximize the data’s potential. Four phases were included in the evaluation: data review, workflow review, performance evaluation, and data usability improvement. (1) The Data Review phase proposes QC metrics and define enhanced metrics for the mobile LiDAR dataset based on current Quality Management Plan for data acquisition. It includes a series of data quantity and quality review to both the LiDAR data and metadata. (2) The Workflow Review phase intends to inspect the characteristics, radiometric and geometric properties and parameterized measurement for target features, examine completeness and integrate of the workflow, assess computation efficiency for key components, etc. (3) The Performance Evaluation phase investigates the opportunity for enhance feature extraction workflow, in terms of enhanced computational infrastructure, algorithm, and so on. (4) The Data Usability Improvement phase is addressed in three aspects: a) Data Repository and Data Management which examines the procedures in managing, accessing and sharing the data, and proposes efficient data repository and data warehousing framework to avoid redundancy and eliminates duplicate efforts in a data management workflow. b) Data Fusion which evaluates the opportunity and feasibility of fuse current mobile LiDAR dataset with other remotely sensed dataset. c) Data Visualization takes advantage of new 3D paradigms and investigate effective web-based, in-browser interactive visualization approaches integrated with 2D and 3D components.