IN-SITU CHANGE DETECTION USING 3D LASER SCANNING

Public highway construction and maintenance are constantly threatened by geologic hazards such as coastal erosion, landslides, seismic loading, etc. Repeat surveys using terrestrial laser scanning (TLS, ground-based LIDAR) enable rapid 3D data acquisition to map, see, analyze, and understand the processes generating such problems. Previously, change detection and analysis between scan surveys was conducted during post-processing upon return to the office, instead of while collecting data in the field, substantially limiting the effectiveness of the field investigation. We have developed a new algorithm that quickly geo-references scans upon field acquisition and simultaneously performs change detection by comparing these newly acquired scans to baseline models.

Testing of the algorithm is underway at several troublesome and unstable sites in Oregon. Two test sites showing active movement along the Highway 101 include Spencer Creek Bridge and the Johnson Creek landslide. An additional trial site is the US20 Pioneer-Eddyville highway realignment project, where several active landslides and surficial slope failures of embankment fill slopes have significantly disrupted construction efforts.

Performing change detection in the field offers several significant advantages to current workflows that require post-processing in the office. First, field change detection serves as an augmented reality system, enabling field crews and researchers to see immediate results and to objectively make key observations while present at the site, without relying exclusively on their personal memories or notes. Second, it can improve the efficiency of the survey. When topographic information is available to the operator during field data acquisition, areas of minimal change can be efficiently surveyed at coarser resolutions and areas of substantial change can be scanned at higher resolutions. This also translates into reduced processing time and data storage requirements, which are currently significant hurdles for analyzing and maintaining 3D laser scan datasets. Finally, this method provides immediate validation and quality control of the RTK GPS and laser scan data being collected, leading to more confidence in the acquired data and allowing immediate resolution of potential issues in the field.