The objective of the Speed Gun App is to empower government officials and transportation advocates. It is well known that speeding, particular in urban areas, is extremely dangerous pedestrians and cyclists. However, concerned citizens are often powerless to tackle these problems. The Speed Gun App will allow users to obtain the approximate speed of passing cars. Its use is not intended for enforcing, but instead for drawing awareness to localized speeding problems.
One does not have to go too far to witness dangerous speeding within densely populated areas. In some cases, the speeding is endemic to a particular location, and even though citizens in the area are aware of it, they are often powerless to draw attention to the problem until a serious accident happens. Matt Bauman’s recent work on determining car speeds from cell phone video [1] was a powerful visual demonstration of how dangerous road conditions can be. To obtain the video for processing, Matt had to create fairly specific conditions. By getting into a high floor of a tall building he was able to obtain a nearly “top-down” view of the scene. This greatly facilitated the image processing necessary to determine the car speeds. Even in this particular beneficial circumstance, Matt’s method still required a lot of manual processing. This project aims to address all these shortcomings and create an easy to use app that can overlay a video with the approximate speeds of the cars in the scene. The only requirement is that the license plate of the cars be visible. The license plate will be used to determine the car’s speed, but will be obfuscated in the final recording, since the objective of the app is not to assign blame to a specific driver, but instead to address dangerous road circumstances that facilitate speeding and initiate a discussion on how the speeding can be mitigated. Specifically, we aim to create an iOS app that will use the OpenCV library for image processing. After a simple calibration procedure, the user will be able to record a short video. In this video the license plates of passing cars will be determined (automatically or with the user intervention, if necessary). By tracking the license plates and using the knowledge that their dimensions are fixed (12” by 6”), we will use Computer Vision methods to determine the approximate car’s speed. A video will be created where these speeds are overlaid. The user will then have the opportunity to submit the video for review. Since user errors (accidental or deliberate) can lead to incorrect speed measurements, all videos will be reviewed before release. We will also conduct discussions with stakeholders before releasing the app to the general public. Note: The original concept for this project was born at an UTC organized event “Traffic21/BikePgh! Speed Monitoring Hack Night”. The PI is currently working on an early concept prototype that may be ready in time for the UTC submission. References: [1] Ed Blazina “Pitt student’s study on Oakland traffic patterns sparks citywide safety effort”. Pittsburgh Post-Gazette. November 19, 2015
Start of month 1 to end of month 3: User interface prototype with video recording capabilities Start of month 3 to end of month 6 (concurrent with above): License plate identification and tracking capabilities Start of month 7 to end of month 10: Computation of approximate speed capabilities Start of month 11 to end of month 12: Back-end for video upload. Demonstrations to stakeholders. Deployment to selected users group. Start of month 7 to end of month 12 (concurrent with above): Collection of test dataset and deployment at key locations to create app awareness.
Initial deployment for collection of testing dataset will be guided by technical needs and include locations close to CMU campus. Further deployments will include key city speeding locations as suggested by stakeholders.
- Creation of a vision-based approximate speed measurement app - Deployment to a select group of users (including bike/ped advocates and city officials) - Collection of key decision-making data for city officials, advocacy groups and other stakeholders - Multiple deployments at different Pittsburgh locations to create app awareness
Name | Affiliation | Role | Position | |
---|---|---|---|---|
bpires@cmu.edu | Pires, Bernardo | Robotics Institute | PI | Faculty - Research/Systems |
Type | Name | Uploaded |
---|---|---|
Progress Report | 48_Progress_Report_2017-09-30 | Sept. 30, 2017, 8:58 p.m. |
Presentation | Speed Gun App: Increasing Awareness of Urban Speeding | March 30, 2018, 11:27 p.m. |
Progress Report | 48_Progress_Report_2018-03-31 | March 30, 2018, 11:27 p.m. |
Final Report | 48_-_Pires_SpeedGunApp_FinalReport_sW79SBf.pdf | Nov. 30, 2018, 6:43 a.m. |
No match sources!
No partners!