5 April 2017 Automated vehicle counting using image processing and machine learning
Author Affiliations +
Abstract
Vehicle counting is used by the government to improve roadways and the flow of traffic, and by private businesses for purposes such as determining the value of locating a new store in an area. A vehicle count can be performed manually or automatically. Manual counting requires an individual to be on-site and tally the traffic electronically or by hand. However, this can lead to miscounts due to factors such as human error A common form of automatic counting involves pneumatic tubes, but pneumatic tubes disrupt traffic during installation and removal, and can be damaged by passing vehicles. Vehicle counting can also be performed via the use of a camera at the count site recording video of the traffic, with counting being performed manually post-recording or using automatic algorithms. This paper presents a low-cost procedure to perform automatic vehicle counting using remote video cameras with an automatic counting algorithm. The procedure would utilize a Raspberry Pi micro-computer to detect when a car is in a lane, and generate an accurate count of vehicle movements. The method utilized in this paper would use background subtraction to process the images and a machine learning algorithm to provide the count. This method avoids fatigue issues that are encountered in manual video counting and prevents the disruption of roadways that occurs when installing pneumatic tubes
© (2017) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Sean Meany, Edward Eskew, Rosana Martinez-Castro, Shinae Jang, "Automated vehicle counting using image processing and machine learning", Proc. SPIE 10170, Health Monitoring of Structural and Biological Systems 2017, 101703G (5 April 2017); doi: 10.1117/12.2261251; https://doi.org/10.1117/12.2261251
PROCEEDINGS
7 PAGES


SHARE
RELATED CONTENT


Back to Top