View Photo(s) of Award Presentation
Project Goal
Douglas County GIS (DCGIS) wanted to extract ADA curb ramps for Douglas County and the City of Omaha using ESRI’s Deep Learning geoprocessing tools and high-resolution imagery.Business Problem Solved
Douglas County and the City of Omaha must abide by the stipulations of the Americans with Disabilities Act (ADA) of 1990. Over the years, while Omaha Public Works has been improving its roadway intersections and adding new intersections, each of those projects included installation of ADA curb ramps. Most of the new ADA curb ramps are added using GIS tools, but there are many that get missed. Creating a Deep Learning model that identifies all ADA curb ramps within Douglas County helped to solve this inventory problem.
Technology Implemented
To accomplish this task, DCGIS created a Deep Learning model that used high resolution aerial orthophotography to detect and extract ADA curb ramps. A Mask Region-based Convolutional Neural Network (Mask RCNN) was trained using 753 ADA curb ramp training samples. After the model was successfully trained, we ran it against our countywide orthophotography.
Development Team Biography
Steve Cacioppo worked on this project. He is the Senior GIS Analyst for DCGIS. Steve has been employed with Douglas County for 20 years.