Abstract: Analysing image sequences to detect and determine temporal events is often known as video analysis. The “video analysis is used in a wide range of domains including entertainment, health care, automotive, transport, home automation, safety and security. Object detection is the first and most important step of moving object tracking. These techniques can be divided into several categories. Moving objects extraction from background is an important task. Results of object extraction depend upon the variation of local or global light intensities, objects shadow, background and foreground regular or irregular movement. Present a robust tracking method by exploiting a fragment-based appearance model with consideration of both temporal continuity and discontinuity information. From the perspective of probability theory, the proposed tracking algorithm can be viewed as a two-step optimization problem. In the first step, by adopting the estimated occlusion state as a prior, the optimal state of the tracked object can be obtained by solving an optimization problem, where the objective function is designed based on the classification score, occlusion prior, and temporal continuity information. In the second step propose a discriminative occlusion model, which exploits both foreground and background information to detect the possible difficult appearance, and also models the consistency of occlusion labels among different frames. In addition, a simple yet effective training strategy is introduced during the model training (and updating) process, with which the effects of spatial-temporal consistency are properly weighted. The propose tracker is evaluated by using the recent benchmark data set, on which the results demonstrate that our tracker performs favourably against other state-of-the-art tracking algorithms.
Keywords: Spatial Reasoning, Visual Surveillance, Temporal Reasoning.