الفهرس 
الغلافOnly 14 pages are availabe for public view
 |  | from | 88 |  |  |
| | | from | 88 | | |
Abstract
Object tracking in video Files is an important topic in computer vision. It has various applications in video summarization, surveillance, robot technology, etc. Every object tracking method requires an object detection mechanism for detecting object either in every frame or when the object first appears in the video.
Object detection is an important computer technology related to computer vision and image processing that deals with detecting objects of a certain class in digital images or videos. Object detection has many applications such in different fields of computer vision, including image retrieval , object tracking, video surveillance, and digital signature.
This thesis purposes object detection module to detect the first appearance of object in accurate location. This model use contour and segmentation method to reduce error rate as much as possible. A method for detecting objects in video files is provided by extracting frames of video. The contour of each object is get using Sobel edge detection operator and fill each connected region. Accordingly, it will be possible to extract objects’ position and other properties. Results showed better performance against other attempts.
The accuracy of the proposed model is 98.377% that is more accurate than the previous models which are 95.6% in ,”2D TARGET TRACKiNG USING KALMAN FILTER”, model that using the background subtraction and 97.5 in ,” Mufti-object Tracking in Video Sequences Based on Background Subtraction and SIFT Feature Matching ”, model that used the combination between background subtraction and SIFT for detection.