Akıllı video gözerimi için hareketli nesne bulma, takip etme ve sınıflandırma

ABSTRACTMOVING OBJECT DETECTION, TRACKING ANDCLASSIFICATION FOR SMART VIDEOSURVEILLANCEYiğithan Dedeoğlug gM.S. in Computer EngineeringSupervisor: Assist. Prof. Dr. Uğur Güdükbayg uuAugust, 2004Video surveillance has long been in use to monitor security sensitive areas suchas banks, department stores, highways, crowded public places and borders. Theadvance in computing power, availability of large-capacity storage devices andhigh speed network infrastructure paved the way for cheaper, multi sensor videosurveillance systems. Traditionally, the video outputs are processed online byhuman operators and are usually saved to tapes for later use only after a forensicevent. The increase in the number of cameras in ordinary surveillance systemsoverloaded both the human operators and the storage devices with high volumesof data and made it infeasible to ensure proper monitoring of sensitive areas forlong times. In order to filter out redundant information generated by an array ofcameras, and increase the response time to forensic events, assisting the humanoperators with identification of important events in video by the use of ?smart?video surveillance systems has become a critical requirement. The making ofvideo surveillance systems ?smart? requires fast, reliable and robust algorithmsfor moving object detection, classification, tracking and activity analysis.In this thesis, a smart visual surveillance system with real-time moving ob-ject detection, classification and tracking capabilities is presented. The systemoperates on both color and gray scale video imagery from a stationary camera.It can handle object detection in indoor and outdoor environments and underchanging illumination conditions. The classification algorithm makes use of theshape of the detected objects and temporal tracking results to successfully cat-egorize objects into pre-defined classes like human, human group and vehicle.The system is also able to detect the natural phenomenon fire in various scenesreliably. The proposed tracking algorithm successfully tracks video objects evenin full occlusion cases. In addition to these, some important needs of a robustiiiivsmart video surveillance system such as removing shadows, detecting sudden il-lumination changes and distinguishing left/removed objects are met.Keywords: Video-Based Smart Surveillance, Moving Object Detection, Back-ground Subtraction, Object Tracking, Silhouette-Based Object Classification,Fire Detection.