The rest of this paper is organized as follows. Section II presents the procedure of the proposedmoving objects detection method. In Section III, experimental results are discussed. Finally, SectionIV presents concluding remarks.The video surveillance market is increasingly moving toward cheaper, efficient, portable, and highresolution systems.a need for better video resolution has generated an interest in using high definition (HD) rates forvideo surveillance.Our prototyped architecture is more hardware-efficient than previous FPGA-based architectures interms of power consumption, area, throughput, and memory utilization.THERE IS AN increasing need for high quality, real-time video on low power, mobile devices suchas mobile phone cameras, bio-medical devices, security and sensor cameras, and so on.Motion estimation (ME) is the most powerful compression tool in the H.264/AVC , but it is alsothe most time consuming. In order to meet low power and real-time constraints, it is critical toefficiently speed up ME via fast algorithms and hardware acceleration
This paper deals with an intelligent image processing method for the video surveillance systems. Wepropose a technology detecting and tracking multiple moving objects, which can be applied toconsumer electronics such as home and business surveillance systems.To detect and track the specific moving objects only, it is important to eliminate the environmentaldisturbances such as light scattering, leaves, birds and so on from input images. To do this, twomethods are mainly used. One is to use the Bayesian method such as the particle filter (PF) or theextended Kalman filter (EKF), the other is to use the difference image methods such as thebackground modeling (BM) or the Gaussian mixture model (GMM).Motion detection is the first essential process in the extraction of information regarding movingobjects and makes use of stabilization in functional areas, such as tracking, classification, recognition,and so on.Our method achieves complete detection of moving objects by involving three significant proposedmodules: a background modeling (BM) module.The three major classes of methods for motion detection are background subtraction, temporaldifferencing, and optical flow . Background subtraction 
,  is the most popularmotion detection method and consists of the differentiation of moving objects from a maintained andupdated background model, which can be further grouped into parametric type and non-parametrictype . Based on the implicit assumption along with the choice of parameters, the parametric modelmay achieve perfect performance corresponding to the real data along with parametric information.The currently implemented method for background subtraction accomplishes its objective bysubtracting each pixel of the incoming video frame from the background model, thus generating anabsolute difference.Although the currently implemented background subtraction method is convenient forimplementation, the noise tolerance in the video frame relies on the determined threshold.Functionalities such as object classification, tracking, behaviour, and identification are then performedon the regions where moving objects have been detected.