Covariance matching-based active contour tracking method

Abstract

The invention relates to a covariance matching-based active contour tracking method and belongs to the technical field of visual tracking. In the covariance matching-based active contour tracking method, an image area energy term is modeled by using non-Euclidean geometry. The method comprises the following steps of: manually initializing a curve surrounding an objective and establishing a covariance matrix as a template of an objective contour for an area surrounded by the curve in a first frame; after the contour of the objective is obtained, recording a level set function value of the template to make preparation for a prior shape and calculating a symbolized distance function of the template; from the image of the next frame, deducing a gradient descent flow from a result of the previous frame according to the established energy functional and updating the level set function; and checking whether iteration stops or not. In the method, the tracking result is more accurate; meanwhile, the covariance matrix is used as an area descriptor and all kinds of information in an image sequence and the correlation between all kinds of information are considered comprehensively, and the method does not depend on foreground and background information distribution, so that the tracking method has universality.

Description

[0001] technical field [0002] The invention relates to an active contour tracking method based on covariance matching, which belongs to the technical field of visual tracking. [0003] Background technique [0004] Moving object tracking is one of the classic topics in the field of computer vision and has important application value. In real-world applications, the tracking problem is recognized as a challenging problem due to poor image quality, environmental lighting changes, shadows, occlusions, and target deformations. [0005] Visual object tracking can be divided into the following categories: point object tracking, kernel tracking and contour tracking. Point target tracking uses one or more points to model the visual target; the kernel-based tracking method has the advantages of simple and fast operation, and has been widely used in visual tracking in recent years, but the target shape expression in nuclear tracking often uses such as ellipse , rectangle and othe...

AI Technical Summary

Problems solved by technology

The advantage of the former is that the computational complexity is low, and the disadvantage is that it does not have topology adaptability. The advantage of the latter is that it has topology adaptability and computational stability, but the disadvantage is that the computational cost is high.

[0007]Terzopoulos defined kinetic energy, potential energy, and damping terms from the principle of Lagrangian dynamics, and derived a dynamic deformation model that unified shape and motion description, expressed as A time-varying and inertial dynamic profile. The equation of motion itself expresses a tracking mechanism based on force balance. The shape constraint is a general smoothness constraint. The disadvantage is that the system dimension is too high and is easily affected by noise; Peterfreund proposed Speed ​​Snake, which organically combines optical flow estimation, active contour model, and Kalman filter, can effectively deal with noise, occlusion, and complex backgrounds to a certain extent. The disadvantage is that the deformation dimension is too high and the calculation is complicated; in order to solve complex For the contour tracking problem of multi-peak observation models under complex backgrounds such as objects, Isard proposed a conditional probability propagation algorithm, that is, the Condensation algorithm. Probability distribution, the generated filter can robustly track the target movement in the complex background. The disadvantage of the Condensation algorithm is: compared with the Kalman filter method, the calculation cost is still too high. If the measurement model modeling is not reasonable, no matter How increasing the number of samples does not improve tracking performance

Images