Automated Localization and Functional Condition Assessment of ADA Curb Ramps With Mobile Lidar Point Clouds
-
2022-01-01
-
Details:
-
Creators:
-
Corporate Creators:
-
Corporate Contributors:
-
Subject/TRT Terms:
-
Publication/ Report Number:
-
Resource Type:
-
Geographical Coverage:
-
Edition:Final Draft
-
Corporate Publisher:
-
Abstract:Curb ramps are an essential component of safe, accessible, and efficient mobility for all transportation users. To make sure that curb ramps can function as intended, their design and construction should meet Americans with Disabilities Act (ADA) standards and guidelines, given that those with disabilities are most adversely affected by improper curb ramp construction. Missing curb ramps, as well as those that do not meet the requirements, may cause accessibility barriers for persons with disabilities. One of the primary challenges that transportation agencies face is that assessing the quality of curb ramps is time-consuming and labor intensive, especially because every corner at an intersection includes multiple curb ramps. Mobile light detection and ranging (lidar) is a remote sensing technology that provides detailed 3D geometry information in the form of 3D point clouds, which can be used to extract various characteristics and metrics to determine the ADA compliance of curb ramps. However, manual processing of mobile lidar data can often be tedious and time-consuming and requires specialized software and training. These issues prevent agencies from using lidar for assessing curb ramp ADA compliance. Therefore, the research team developed an automatic workflow process to extract and localize curb ramps within a large data point cloud. The proposed approach consists of three steps: ground filtering, curb detection, and curb ramp localization. It was evaluated both qualitatively and quantitatively with a mobile lidar data set. The recall, precision, and F-1 scores were all 72.4 percent. The proposed approach can be potentially used for further analysis, such as feature characterization and point cloud classification of other features.
-
Format:
-
Funding:
-
Collection(s):
-
Main Document Checksum:
-
Download URL:
-
File Type: