An accurate inventory of unpaved road network length and condition within a county, state, or region is important for efficient use of resources to manage and maintain this critical transportation asset. Object-based classification techniques provide a cost-effective way to identify unpaved roads within a local agency's road network when the road type (i.e., paved versus unpaved) attribute is missing. We present a Trimble eCognition® algorithm using four band optical aerial imagery and object-based classification to classify roads as paved or unpaved. The ruleset evaluates relationships between bands and allows separation and segmentation of unpaved roads from other pavement classes. The algorithm is applied to unincorporated areas of a six county region in Southeastern Michigan. Tree shadows on roads and the spectral similarity of road construction materials pose challenges to classification accuracy. An accuracy assessment of the classification indicated that the algorithm works well with overall classification accuracy between 82 and 94 percent.

The need of local governments and transportation agencies to periodically asses the condition of unpaved roads in a cost-effective manner with rapid response times has lead to interest in the use of UAVs (Unmanned Aerial Vehicles) and remote sensing technologies. Currently these assessments are done through visual inspections with agency staff making occasional spot measurements. An unpaved road assessment system was developed to address these issues while at the same time providing a more accurate means of characterizing distresses and determining the roads condition for inspectors. This system uses a single-rotor UAV with a Digital Single-lens Reflex (DSLR) camera to capture overlapping imagery of unpaved roads. The UAV is equipped with a full combination GPS plus IMU (Inertial Measurement Unit) that allows it to fly predetermined waypoints with great stability while at the same time allowing the pilot the ability to take over at any time. Collected imagery is analyzed to locate road distresses. The imagery is run through a Structure From Motion (SfM) algorithm that generates a 3D model of the road surface from which additional condition information can be characterized. This system is easily transported and rapidly deployable to sections of unpaved roads for assessment.