An Efficient Method for Moving Vehicle Detection in Real-Time Video Surveillance
Surveillance systems are important to monitor and control critical situations for the field of traffic flow. Our aim is to enhance the ITS system about highway traffic and retrograde vehicle detection from surveillance video. We review the literature abou
- PDF / 579,193 Bytes
- 9 Pages / 439.37 x 666.142 pts Page_size
- 90 Downloads / 201 Views
Abstract Surveillance systems are important to monitor and control critical situations for the field of traffic flow. Our aim is to enhance the ITS system about highway traffic and retrograde vehicle detection from surveillance video. We review the literature about vehicle detection in various countries and major highways, day to day vehicle transportation face critical situations then instant accident from different locations. This paper proposes an efficient method of detecting traffic surveillance in moving vehicles for various road conditions. Kalman filtering predicts the area where to search for each corner feature. This system is used to detect moving vehicles effectively in complicated situations such as pedestrians crossing, weather conditions, roadside trees, etc. The tracking algorithms are carried out over a various range of vehicles. It is to determine the regions where significant motion has occurred. Keywords Intelligent transport system · Traffic flow · Video surveillance
1 Introduction The forthcoming technological advances in various fields are growing rapidly for vehicle transportation, namely Intelligent Transport System (ITS). The method of traffic arrangements to the road-based transport system has been applied to the developed countries and developing countries [1]. This system uses CCTV Cameras for the purpose of surveillance. The vehicle transportation and decision-making are used to calculate the number of vehicles. This task mainly depends on foreground extraction. Normally the natural background includes detecting large objects such as buildings, plants, roads, apartments, etc., it has similar intensity values but intensities differ considerably with each other [2]. The foreground method is used to extract the S. Sri Jamiya (B) · P. Esther Rani Electronics and Communication Engineering, Vel Tech Rangarajan Dr. Sagunthala R&D Institute of Science and Technology, Avadi, Chennai, India e-mail: [email protected] P. Esther Rani e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2021 P. Suresh et al. (eds.), Advances in Smart System Technologies, Advances in Intelligent Systems and Computing 1163, https://doi.org/10.1007/978-981-15-5029-4_47
577
578
S. Sri Jamiya and P. Esther Rani
background image which captures from optical flow model. It carries out a separate foreground from a background [3] by learning and capture of the vehicle background image. This algorithm is used to detect particular moving objects to make robustness and also various changes have been considered to detect abnormal behavior of a vehicle [4]. Depending upon the frame-level and pixel-level specification, true positive detection rate and false-positive detection rates are calculated by detecting abnormally moving vehicles. It is considered to be positive when detected events are abnormal and are said to be negative when detected events are normal. The incremental framework of background modeling is relatively closer to our proposed work and the number of vehicles is counted [5]. The opti
Data Loading...
