Object extraction from real time video image and an automated object tracking application

Real time video processing is known as composition of reliable techniques for applications such as national border security, traffic monitoring, event detection and public security which requires high reliability and correctness rates. The objective of this study is to track moving objects in real time video images. Background subtraction techniques are applied on the real time video images in order to detect lines and borders of moving objects. Image threshold techniques are applied onto background subtraction result and pixel values which are below the image threshold are omitted to increase the efficiency of the developed algorithm. In this study it is observed that background subtraction and image threshold techniques generate satisfactory results for moving object detection in real time video images. Occlusion handling is one of the key purposes of the study, special movement algorithms are used for occlusion handling such as movement template generation. Movement template generation is used to handle occlusion by generating movement template; position estimation of the object can be done when the moving object is occluded by an obstacle or become invisible from the camera?s view. It is also observed that it is possible to generate successful outputs fro tracking moving objects in the video frames, even when the moving object is partially or entirely occluded by an obstacle. The classification results and the values of statistical parameters indicated that the combination of background subtraction and movement template generation for moving objects results feasible outputs for real time video object tracking.Keywords: Object Tracking, Image Threshold, Background Subtraction, Movement pattern.