Details:
-
Creators:
-
Corporate Creators:
-
Corporate Contributors:
-
Subject/TRT Terms:
-
Publication/ Report Number:
-
Resource Type:
-
Geographical Coverage:
-
Edition:Technical Report 151
-
Corporate Publisher:
-
Abstract:In this project, we refined an approach to using data collected by the City of Austin’s traffic monitoring cameras to automatically identify pedestrian activities on roadways. Our approach automatically analyzes the content of video data gathered by existing traffic cameras using a semi-automated processing pipeline powered by state-of-art computing hardware and algorithms. The method also extracts a background image at analyzed locations, which is used to visualize locations where pedestrians are present and display their trajectories. We illustrate the use of a scalable tool for the automated analysis of data collected from monocular traffic cameras that can allow agencies to leverage existing infrastructure in the analysis and mitigation of pedestrian safety concerns with two use cases. This work focuses on the following two major goals: 1) evaluation of pedestrian activities before and after a pedestrian hybrid beacon device installation; and 2) assessment of pedestrian activities at new locations. The specific activities can be further grouped into four tasks: Task 1: Pedestrian crossing detection before and after PHB installation at Anderson location; Task 2: Recording and analysis of pedestrian crossing event at Rundberg location; Task 3: Recording and analysis of pedestrian crossing at Payton location; and, Task 4: Bus stop usage inferences at Payton location. In this report, we document our efforts, findings, challenges, and achievements in this project.
-
Format:
-
Funding:
-
Collection(s):
-
Main Document Checksum:
-
Download URL:
-
File Type: