Object detection in surveillance videos

摘要: In this thesis, a novel scheme for object detection in complex background scenes has been proposed.The input videos used have fixed backgrounds and static cameras. Initially median of few frames is evaluated obtaining proper estimate the background.Local threshold based subtraction done extracting objects from video sequence.During sudden illumination changes, optical flow analysis motion segmentation.It assumed that during photometric distortions, motion.Subsequently shadow suppression to resulting thresholded image. Hue Saturation Value(HSV) color space model suppression.Visual measures convey performance algorithm.