Survey on Intruder Detection from Surveillance Videos
Palagati Harish1, K.Priya2
1Palagati Harish, PG Scholar Dept. of Information Technology Sathyabama University Chennai, India.
2K. Priya, Assistant Professor Faculty of Computing Sathyabama University Chennai, India.
Manuscript received on January 25, 2014. | Revised Manuscript received on February 13, 2014. | Manuscript published on February 28, 2014. | PP: 177-178 | Volume-3, Issue-3, February 2014. | Retrieval Number: C2639023314/2013©BEIESP
Open Access | Ethics and Policies | Cite
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Surveillance cameras are the main source to detect malpractices, thefts, intruders in our daily life. These cameras records large volumes of data in video format. The data recorded by surveillance cameras are just stored and they are not processed. Several methods are introduced to track and extract the features from these videos to detect intruder. But, the detection will mismatch in crowded areas. Detection of human activity is complex in such areas. Semantic Content Mining (SCM) is proposed to extract the semantic content like objects and events from the video. This method is useful for the detection of human activity and his behavior. Construction of ontology provides spatial and temporal relations between different video frames. This provides the events or actions performed by intruders.
Keywords: Semantic content, Ontology; Human activity recognition, Spatial and Temporal Relations.