MOVING OBJECT DETECTION USING BIT PLANE SLICING

Abstract

This thesis presents moving object detection algorithm using bit plane extraction of successive frames and comparing the respective bit planes by XOR operation. The proposed method works on 8-bit grayscale video frames obtained from a static camera. This algorithm is able to detect the motion of single and multiple objects in outside and inside environments. Algorithm has been implemented in MATLAB by using several videos from VISOR database and was compared to existing conventional methods to show its effectiveness. Performance of an algorithm was evaluated based on ground truth metrics and results in terms of sensitivity, specificity, positive prediction and accuracy proved the validity of it. Results show that the proposed algorithm performs better in terms of mentioned metrics in comparison to other algorithms.