Moving Vehicles Detection Based on Improved Gaussian Mixture Model

Abstract

Vehicles detection has a great significance on traffic congestion relief, traffic accidents prevention and treatment. This paper proposes an improved Gaussianmixture model(GMM) to detect moving vehicles because GMM always detects background for foreground in complex environment. The new algorithm combines GMM with frame difference. Then, the whole vehicle can be detected by this new algorithm because of the enhanced demarcation point between background and foreground. Finally, the moving vehicles detection system is established. Experiments in different weather(sunny and rainy) and different time(day and night) have been made. They can prove that the new algorithm has been greatly improved in the aspects of adaptability, accuracy, real-time and so on. The movingvehicles can also be detected correctly and effectively in the situation with various complicated factors.

Original document

The different versions of the original document can be found in:

• http://dx.doi.org/10.2991/eame-15.2015.210 under the license https://creativecommons.org/licenses/by-nc

• https://download.atlantis-press.com/article/22422.pdf

• https://www.atlantis-press.com/proceedings/eame-15/22422,

http://or.nsfc.gov.cn/handle/00001903-5/160637,

https://academic.microsoft.com/#/detail/1950212973 under the license cc-by-nc