抽象的

Abandoned Object Detection for Automated video Surveillance using Hadoop

Tejas Naren TN, Shankar SiddharthKA , Venkat Krishnan S,Sanjeevi LR, Dr. S Chandrakala

In the recent years, preventive security measures at public places has gained lot of importance. Visual Surveillance has now got huge attention to ensure safety in public places. These systems generate huge amount of data on a daily basis. This data has enormous Volume, Velocity ,Variety. Hence , This is can be termed as Big Data. Handling Big Data is a challenge and thus, in this proposed system , the processing of the video is done on Hadoop. The existing system takes more time to process these live stream video data which is inversely proportional to what we except.The proposed system process these data in an efficient and a faster manner in a parallel distributed environment .Nowadays, the pure mass of information that has to be handled by the operators of such systems has over- grown their capabilities. Hence , A System capable of identifying objects that tend be vulnerable to the surrounding environment is addressed in the system. This System also implements Machine Learning to overcome the common technical problems present in the existing system which includes lighting changes, occlusions ,cluttered backgrounds.