N. Al-Yamani, S. Qaisar, Abrar S. Alhazmi, S. Mohammad, A. Subasi
{"title":"An event driven surveillance system","authors":"N. Al-Yamani, S. Qaisar, Abrar S. Alhazmi, S. Mohammad, A. Subasi","doi":"10.1109/ICEDSA.2016.7818561","DOIUrl":null,"url":null,"abstract":"In recent era surveillance systems play a vital role in our daily life. The goal of surveillance systems is to collect the desired information about concerned targets in the sensing environment. Traditional surveillance systems are limited by a predefined monitoring pattern. It can render a reduced system reliability. Moreover, the system is always functional regardless of the target presence. It results in an increased and useless system activity, resources utilization and power consumption. In this context an event driven human surveillance system is devised. It overcomes these drawbacks up to a certain extent by smartly combining a sensors assembly along with a camera, embedded processor and classifiers. Except the low power object detector, detection of concerned object in the surveillance zone, the remaining system is kept in sleep mode. On the detection of an event the system modules like embedded processor, target location detectors, camera, etc. are activated. The event occurrence can also activate the lightning system depending on the sensing environment illumination condition. The target location, within the field of view, is provided by the location detectors. This information is employed to pilot the actuator in order to update the camera positioning. The captured images are analysed by the human detector module. Further actions like image storage, video recording, etc. are taken depending on the human detector output. The proposed system functionality is verified with an experimental setup. Results demonstrate the devised system self-organisation feature. It results into an improved performance of the proposed system, in terms of system resources utilization, reliability and power consumption compared to the counter traditional ones.","PeriodicalId":247318,"journal":{"name":"2016 5th International Conference on Electronic Devices, Systems and Applications (ICEDSA)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2016-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"12","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2016 5th International Conference on Electronic Devices, Systems and Applications (ICEDSA)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICEDSA.2016.7818561","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 12
Abstract
In recent era surveillance systems play a vital role in our daily life. The goal of surveillance systems is to collect the desired information about concerned targets in the sensing environment. Traditional surveillance systems are limited by a predefined monitoring pattern. It can render a reduced system reliability. Moreover, the system is always functional regardless of the target presence. It results in an increased and useless system activity, resources utilization and power consumption. In this context an event driven human surveillance system is devised. It overcomes these drawbacks up to a certain extent by smartly combining a sensors assembly along with a camera, embedded processor and classifiers. Except the low power object detector, detection of concerned object in the surveillance zone, the remaining system is kept in sleep mode. On the detection of an event the system modules like embedded processor, target location detectors, camera, etc. are activated. The event occurrence can also activate the lightning system depending on the sensing environment illumination condition. The target location, within the field of view, is provided by the location detectors. This information is employed to pilot the actuator in order to update the camera positioning. The captured images are analysed by the human detector module. Further actions like image storage, video recording, etc. are taken depending on the human detector output. The proposed system functionality is verified with an experimental setup. Results demonstrate the devised system self-organisation feature. It results into an improved performance of the proposed system, in terms of system resources utilization, reliability and power consumption compared to the counter traditional ones.