Event Analysis in Intelligent Aerial Surveillance Systems for Vehicle Detection and Tracking
B.T.R. Naresh Reddy1, Prasad Nagelli2, K. Srinivasulu Reddy3
1B.T.R. Naresh Reddy,  Research Scholar, Mewar University, Chittorgarh, Rajasthan. India.
2Prasad Nagelli, Research Scholar, JNT  University, Hyderabad, (A.P), India.
3K. Srinivasulu Reddy, Computer Science & Engineering, JNTU University, Vardhaman College Of Engineering, Hyderabad, (A.P), India.
Manuscript received on September 24, 2013. | Revised Manuscript received on October 16, 2013. | Manuscript published on October 30, 2013. | PP: 262-270  | Volume-3, Issue-1, October 2013. | Retrieval Number:  A2251103113/2013©BEIESP

Open Access | Ethics and Policies | Cite
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)

Abstract: Vehicle detection plays an important role in the traffic control at signalized intersections. One of the Advanced Event Assistance systems are being researched nowadays for Intelligent Vehicles has to deal with the detection and tracking of other vehicles. The present system to detect and track moving vehicles based on detectors and classifiers. In previous approach escapes some of the existing frameworks for detection vehicles in traffic monitoring systems. Moving vehicles detection based on the pixelwise classification in both detectors and classifiers using multilayer perceptrons and Dynamic bayesian network. Pixel wise classification provides not only region wise but also sliding window also detected the vehicles. The feature extraction performed in both training and detection stages. In the classification used dynamic Bayesian networks and in this network vehicle and non vehicle are identification purpose use a support vector machine. The classification of vehicles and non vehicles are identification purpose used a color histogram algorithm. In this framework used two detectors and two classifiers. Two detectors for local feature extraction are Harris corner detector and canny edge detector. Then, two classifiers of color feature extraction, SVM and multilayer perceptrons are introduced. Both of them have good performance on vehicle color classification but we choice SVM for color feature extraction in our system. Finally, the training process and classification process of dynamic Bayesian network are utilized. In experimental results are shown in different videos are taken at different cameras and different heights in surveillance systems.
Keywords: Aerial surveillance, Canny edge detection, Dynamic Bayesian Networks, Multilayer Perceptrons Soft Computing, Vehicle Detection.