Open AccessForeground Moving Object Detection using Support Vector Machine (SVM)
Author(s) -
M. Nagaraju*,
M. Baburao
Publication year - 2019
Publication title -
international journal of recent technology and engineering
Language(s) - English
Resource type - Journals
ISSN - 2277-3878
DOI - 10.35940/ijrte.d8265.118419
Subject(s) - artificial intelligence , computer vision , viola–jones object detection framework , computer science , video tracking , object (grammar) , object detection , support vector machine , canny edge detector , preprocessor , object class detection , edge detection , frame (networking) , enhanced data rates for gsm evolution , optical flow , process (computing) , pattern recognition (psychology) , image processing , image (mathematics) , face detection , facial recognition system , telecommunications , operating system
Detect the existence of an object and to locate it in a video is called object detection. In the process of tracking an object we first segment the given frame and then track its position, motion and occurrence. Before tracking any object the process of object detection and object classification are done. Once we detect any object we have to classify the object into several categories like humans, vehicles, animals etc. There are many applications in which object tracking is being used such as robotics, traffic monitoring, security, video surveillance and animations. In this paper, we proposed a framework to detect the foreground moving object in a video scene at real time. Firstly preprocessing the given video is divided into frames to detect the object. Next morphological filter is used to remove the Nosie after detect the object. In this step, frame is divided into pixels. Here optical flow method is used to segment the given frames. After that, canny edge detector is used to detect the edge of object from segmented image. Finally, classify the object using Support Vector Machine (SVM)