Tracking method based on dual-model adaptive kernel correlation filtering

Abstract

The invention provides a tracking method based on dual-model adaptive kernel correlation filtering, which comprises the following steps: initializing the position of a pre-estimated target, calculating a Gaussian tag, and establishing a main feature model and an auxiliary feature model; extracting HOG features to serve as features of a main feature model, extracting deep convolution features to serve as features of an auxiliary feature model, and setting initialization parameters; calculating a response layer of the pre-estimated target by utilizing the main characteristic model, and obtainingan optimal position and an optimal scale of the pre-estimated target by the response layer through a Newton iteration method; if the maximum confidence response value max of the response layer corresponding to the optimal scale is greater than an empirical threshold u, determining a pre-estimated target position, and updating the main feature model; if max is smaller than or equal to an empiricalthreshold u, stopping updating the main feature model, expanding a search area, extracting CNN features of a target pre-selected area, performing dimensionality reduction on deep CNN features by using a PCA technology, estimating a new target position by using the dimensionality-reduced CNN features, and updating an auxiliary feature model until thevideo sequence ends.

Description

technical field [0001] The present invention relates to the technical field of computer vision, in particular, to a dual-model adaptive kernel correlation filter tracking method. Background technique [0002] Object tracking is a basic component of the field of computer vision and plays an important role in many practical applications, such as intelligent transportation, intelligent monitoring, etc. Target tracking is to track without any prior knowledge. In the initial state, the target size and target position are given in the first frame of the video image sequence, and the trajectory and target size of the given target are predicted in the subsequent image sequence. . Tracking algorithms are divided into generative tracking algorithms and discriminative tracking algorithms. The generative tracking algorithm is to extract the effective information of the target appearance model in the current frame through certain statistical means (sparse expression, CN, color histogra...

AI Technical Summary

Problems solved by technology

[0003] The reason why the target tracking problem is complicated is that the appearance of the target may be caused by fast movement, complex background, motion blur, deformation, illumination change, rotation inside and outside the plane, low resolution, occlusion, out of view, scale change and other interference factors during the tracking process. The model has changed a lot, so target tracking is still a very challenging problem at present

Images