LITERATURE SURVEY On Moving Object Detection.docx

LITERATURE SURVEY 1. “Real-Time Object Detection and Tracking on a Moving Camera Platform” In this paper, we proposed a real-time visual tracking system on a controlled pan-tilt camera. The input/output HMM is employed to model the overall visual tracking system in the spherical camera platform coordinate. In order to fast detecting and tracking targets on a moving camera at the same time, we adopt the optical flow to observe the different displacement in the image sequence. A two layer visual tracking architecture is proposed to improve the tracking robustness. The bottom level uses the optical flow estimation again for tracking the feature points across image frames. The top level utilizes the tracking result of bottom level and applies the particle filter for estimating the target state.

2. “A Moving Objects Detection Algorithm Based on Improved Background Subtraction” In this material, a new efficient moving target detection method which is a improved background subtraction was proposed to detect moving objects, I summed up two significant advantages, one of them is improved the background subtraction and increased algorithm's running efficiency. Another one is to offset sensitive deficiency of the light changes.

3. “Independent Component Analysis-Based Background Subtraction for Indoor Surveillance” Background subtraction is a widely used approach for detecting foreground objects in videos from a static camera. Indoor surveillance applications such as home-care and health-care monitoring, a motionless person should not be a part of the background. A reference background image without moving objects is, therefore, required for such applications. In this paper, we have presented an ICA-based background subtraction scheme for foreground segmentation. The proposed ICA model is based on the direct measurement of statistical independency that minimizes the difference between the joint PDF and the product of marginal PDFs, in which the probabilities are simply estimated from the relative frequency distributions. The proposed ICA model well performs the separation of highly-correlated signals.

4. “Tracking of a Moving Object Using One-Dimensional Optical Flow with a Rotating Observer” We proposed the moving object tracking method using a one-dimensional optical _ow under the rotating observer. We avoid an ill-posed problem for calculating the optical _ow based on the gradient method using one-dimensional brightness data. In addition, we overcome the restriction of the object motion by spanning the calculation axis along the moving direction of the object to be tracked on both sides of the object. In order to eliminate the apparent motion of the stationary object environment, we introduce the detection method of the moving object by mapping the motion of the stationary environment object into LTS and by the block gradient method. The tracking problem with a rotating observer is regarded as a problem with a _xed observer using pixels, which belongs to the moving object for calculating 1D optical _ow. Hence, we can eliminate the apparent motion of the stationary environment object. 5. “Object Focused Simultaneous Estimation of Optical Flow and State Dynamics” In this paper we examine the convergence of the SEOS optical flow method. Convergence of SEOS was evaluated for both the Gauss-Seidel and Jacobi iterative techniques. It was shown that SEOS will converge for both Gauss-Seidel and Jacobi iterative schemes for any initial approximation. We also examined the effect of applying SEOS to a real-image sequence and tracking various objects within the „Hamburg Taxi Sequence‟. Our results show a uniform directional flow field for the tracked vehicle. All objects outside the vehicle of interest were able to be smoothed out. This object isolation feature of the SEOS technique could prove invaluable in numerous tracking applications. 6. “Fast Background Subtraction and Shadow Elimination Using Improved Gaussian Mixture Model” Background subtraction is an active researching field, because it can be used in many applications. An efficient background subtraction approach is the base which determines performance of the whole system. In this paper, we tried to simplify the original GMM to improve its performance.