Detectıon of Abnormal Activity at College Entrance Through Video Surveillance

Abstract

AbstractMost of the colleges are having the facility to park vehicles for students and staff members in distinct locations. As the location of parking inside campus has been assigned by the college authority in the restricted areas, but some abnormal activity is found related to vehicle parking. So the main objective of this paper is to detect students who park their vehicles inside college or prohibited area. The sample video frames are used and the extracted image from the video frame can be further utilized to extract information of color, logo and number plate recognition. The vehicle number then add to database for the identification of student vehicle for further punishment. Different algorithms, i.e., CNN, ANN, RNN are tested on sample dataset for blue color, logo and number plate detection and CNN found satisfactory result in comparison ANN and RNN.KeywordsCNNANNRNNExtractionSegmentationRecognition