Adaptive object feature tracking method

Abstract

The invention provides an adaptive object feature tracking method, belongs to the technical field of computer graph and image processing and can effectively solve the problem of stably tracking rigid and non-rigid target objects in real time for a long time. The adaptive object feature tracking method comprises the following steps: constructing a pixel-based target grayscale table saving grayscale and pixel point information, wherein the grayscale is contributive to a predicted target center position; performing adaptive two-point feature selection based on the target grayscale table and mutual information so as to determine a feature combination which can perfectly describe a target and model the target. In a tracking process, according to the adaptive object feature tracking method, firstly a Hough table corresponding to images is constructed according to the target grayscale table, secondly a predicted target position is searched and determined according to a target model, thirdly a position corresponding to the maximum pixel point number in the Hough table is searched within a relatively small local range by taking the predicted target position as the center, and the position is taken as the final target position, so that the target positioning is completed. The adaptive object feature tracking method is mainly applied to tracking of object targets.

Description

technical field [0001] The invention belongs to the technical field of computer vision object tracking, in particular to the technical field of computer graphic image processing. Background technique [0002] Visual object tracking is a basic and critical problem for many computer vision applications, such as video analysis, intelligent surveillance, human-computer interaction, behavior recognition, etc. Although researchers have made a lot of work on this, it is necessary to achieve real-time Stable object tracking remains an extremely challenging task. [0003] Since the tracking method based on online learning is inherently more adaptable to changes in the object and its environment, it is beneficial to complete long-term tracking tasks, so the current object tracking methods that rely on detection or learning (such as TLD, Tracking-Learning-Detection) are getting more and more attention. increasingly widespread attention. These methods explore unknown data and informat...

AI Technical Summary

Problems solved by technology

Fan et al. proposed to learn regions of interest with strong discrimination to assist in tracking. However, when there is violent movement in the scene, the computational efficiency is still not high due to the limitations of these local regions.

Godec et al. achieved a satisfactory tracking effect by classifying the background into multiple virtual types by clustering the scene, but this method assumes that the background is only gradually and slightly changed, which is not true in many tracking occasions, so its limited application

Images