Target tracking template updating method

Abstract

The invention discloses a target tracking template updating method. The method comprises the following steps: determining whether a target tracking template needs to be updated; extracting feature points of a target region in the last frame and a current frame of an image, matching the feature points of the target region of the last frame with the feature points of the target region of the current frame, and re-determining the position of a tracked target; using the feature points of the target region of the current frame as new feature points to carry out tracking; and using the target region of the current frame as an updated target tracking template. The beneficial effects are as follows: setting of determination conditions of target tracking template updating avoids the situation that the template cannot catch up with the changed target and prevents noise accumulation during tracking, so that the target tracking is continuous and stable; establishment of an image Gauss pyramid improves the update efficiency; a quick feature point detection method enhances the repeatability of feature detection and prevents unstable edge responses; and the method is suitable for tracking a changing target, especially when wide-angle overturn happens.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to a method for updating a target tracking template. Background technique [0002] Target tracking is to determine the position and attitude of the target in each frame according to the correlation of video information in space and time in the sequence of images, which is widely used in many fields such as intelligent human-computer interaction, medical diagnosis, video surveillance, missile guidance, etc. . Due to the diversity of tracking target forms and the uncertainty of target feature changes, how to achieve effective target tracking in various environments has always been a research hotspot. [0003] The tracking based on the template matching method is a method of pixel-by-pixel comparison. The basic idea is to use the known object as a template to translate on the collected image, and calculate the correlation value between the object and the sub-image. The maximu...

AI Technical Summary

Problems solved by technology

[0004] When the target is close to or far away from the camera, or the camera field of view changes, the target will become larger or smaller, and the size of the template remains unchanged, which may cause two problems: one is that the target becomes smaller, and the template will introduce more than the target. background pixels; second, when the target becomes larger, only a small part of the target exists in the template, and there will not be enough target pixels to distinguish the target from the background

[0005] The update mechanism of the template is a very important technical key, which will directly affect the target tracking performance: if the update frequency is too low, the template cannot keep up with the changes of the target; if the update is too frequent, it is easily affected by noise and introduces errors, and The error is continuously accumulated, which leads to the drift of the tracking window

Images