Showing papers on "Horn–Schunck method published in 2006"

Robust Optical Flow with Combined Lucas-Kanade/Horn-Schunck and Automatic Neighborhood Selection

Abstract: Differential optical flow methods are widely used within the computer vision community. They are classified as being either local, as in the Lucas-Kanade method, or global, such as in the Horn-Schunck technique. Local differential techniques are known to have robustness under noise, whilst global techniques are able to produce dense optical flow fields. We will show that the Horn-Schunck Technique, when combined with Lucas-Kanade, can yield the advantage of having both robust and dense optical flow fields. Selection of neighborhood size is an important tuning parameter for the combined Lucas-Kanade/Horn-Schunck technique. Choosing the optimal neighborhood is a difficult task and greatly effects the performance of optical flow results. We outline a method for the automatic selection of neighborhood size based on Stein's unbiased risk estimator (SURE). Algorithms are derived for a combined Lucas-Kanade/Horn-Schunck technique with automatic neighborhood selection. The performance of SURE neighborhood selection for the combined optical flow technique is simulated via Matlab, providing an illustration of the performance that is attainable.

An Algorithm for Optical Flow Computation Based on a Quasi-Interpolant Operator

Abstract: A fundamental problem in the processing of image sequences is the computation of the velocity field of the apparent motion of brightness patterns usually referred to optical flow. In this paper a novel optical flow estimator based on a bivariate quasi-interpolant operator is presented. Namely, a non linear minimizing technique has been employed to compute the velocity vectors by modeling the flow field with a 2D quasi-interpolant operator based on centered cardinal B-spline functions. In this way an efficient computational scheme for optical flow estimate is provided. In addition the large solving linear systems involved in the process are sparse. Experiments on several image sequences have been carried out in order to investigate the performance of the optical flow estimator.

An Improved Optical Flow Algorithm

Abstract: Optical flow estimation is an important method to motion image analysis.This paper introduces forward and backward constraint equation and Hessian matrix for the computation of optical flow.It examines well-posedness of each point of local neighbourhood and the weight of Lucas-Kanade’s method is defined as the reciprocal of the conditioning number of its Hessian Matrix.This can eliminate those uncertainty constrains and improve the numerical stability of the solution of the gradient constraint equation.Experimental results show that this method is suitable and reliable.

A MPBSG Technique Based Parallel Dual-Type Method Used for Solving Distributed Optimal Power Flow Problems

Abstract: In this paper, we propose a method to solve the distributed optimal power flow problem and discuss the associated implementation. We have combined this method with a projected Jacobi (PJ) method and a modified parallel block scaled gradient (MPBSG) method possessing decomposition effects. With the decomposition, our method can be parallel processed and is computationally efficient. We have tested our method for distributed OPF problems on numerous power systems. As seen from the simulation results, our method achieved a dramatic speed-up ratio compared with the commercial IMSL subroutines.

A parallel algorithm for nonlinear multi-commodity network flow problem

Abstract: In this paper, we propose a method to solve the Nonlinear Multicommodity Network Flow (NMNF) Problem. We have combined this method with a projected-Jacobi (PJ) method and a duality based method possessing decomposition effects. With the decomposition, our method can be parallel processed and is computationally efficient. We have tested our method on several examples of NMNF problem and obtained some successful results.

A Fast Modified Horn & Schunck Method

Abstract: This paper presents a fast , accurate and reliable modified Horn & Schunck approach for robust boundary preserving estimation of optical flow where only 30 iterations are used . The proposed method is derived from the benchmark algorithm of Horn & Schunck and Simoncelli's matched-pair 5 tap filters, such that it produces robust, fast and exact detection of motion boundaries and it is very simple to implement. Experimental results using synthetic and real optical flows are presented to demonstrate the effectiveness of our method in comparison to selected methods .

Voting method for stable range optical flow computation

Abstract: For the non-invasive imaging of moving organs, in this paper, we develop statistically accurate methods for the computation of optical flow. We formalise the linear flow field detection as a model-fitting problem which is solved by the least squares method. Then, we show random-ssampling-and-voting method for the computation of optical flow as model-fitting problem. We show some numerical examples which shows the performance of our method.

Voting method for stable range optical flow computation

Abstract: For the non-invasive imaging of moving organs, in this paper, we develop statistically accurate methods for the computation of optical flow. We formalise the linear flow field detection as a model-fitting problem which is solved by the least squares method. Then, we show random-ssampling-and-voting method for the computation of optical flow as model-fitting problem. We show some numerical examples which shows the performance of our method.

Motion Estimation in Gated Cardiac Emission Tomography by Optical Flow Techniques

Abstract: The purpose of this work was to develop and test a new motion estimation algorithm for gated cardiac emission tomography that is designed to be more accurate than conventional methods particularly for relatively large frame-to-frame displacements. The proposed method consists of a sequential application of the conventional Horn-Schunck algorithm to quadratic approximations of the standard, non-rigid frame registration model (with a quadratic smoothing term). The method was tested on frames reconstructed from noisy and noise-free data generated from the 4D cardiac phantom. Our simulations indicated that the method improved motion estimation accuracy relative to the Horn-Schunck method, with increased improvement at lower noise levels.