期刊名称：International Journal of Innovative Research in Computer and Communication Engineering
印刷版ISSN：2320-9798
电子版ISSN：2320-9801
出版年度：2015
卷号：3
期号：3
DOI：10.15680/ijircce.2015.0303202
出版社：S&S Publications
摘要：Machine-controlled motion detection technology is Associate in nursing integral element of intelligenttransportation systems, and is especially essential for management of traffic and maintenance of traffic policeinvestigation systems. Traffic police investigation systems mistreatment video communication over real-worldnetworks with restricted information measure typically encounter difficulties attributable to network congestion and/orunstable information measure. This is often particularly problematic in wireless video communication. This hasnecessitated the event of a rate management theme that alters the bit-rate to match the procurable network informationmeasure, thereby manufacturing variable bit-rate video streams. However, complete and correct detection of movingobjects beneath variable bit-rate video streams could be a terribly tough task. During this paper, we tend to proposeAssociate in nursing approach for motion detection that utilizes Associate in nursing analysis - primarily based radialbasis perform network as its principal element. This approach is applicable not solely in high bit-rate video streams,however in low bit- rate video streams, as well. The planned approach consists of a varied background generation stageand a moving object detection stage. Throughout the assorted background generation stage, the lower-dimensionalEigen-patterns and also the adjustive background model are established in variable bit -rate video streams bymistreatment the planned approach so as to accommodate the properties of variable bit-rate video streams. Throughoutthe moving object detection stage, moving objects are extracted via the planned approach in each low bit -rate and highbit- rate video streams; detection results are then generated through the output worth of the planned approach. Thedetection results created through our approach indicate it to be extremely effective in variable bit-rate video streamsover real-world restricted information measure networks. Additionally, the planned methodology will be simplyachieved for period of time application. Quantitative and qualitative evaluations demonstrate that it offers blessingsover different state- of-the -art ways. For example, and accuracy rates created via the planned approach were up toeighty six.38% and 89.88% beyond those created via different compared ways.
关键词：Intelligent transportation systems; moving object detection; neural network; principal element analysis;variable bit-rate.
