Single moving target tracking method and device based on fixed form under dynamic background

Abstract

The invention discloses a single moving target tracking method and a device based on a fixed shape under a dynamic background. The method comprises steps: two black cross lines of a white background serve as a to-be-tracked moving target, and a YUV image in the current frame for the to-be-tracked moving target is acquired through a camera; a frame difference method is used for detecting a moving target area in the YUV image; the moving target in the moving target area is subjected to edge detection to acquire a binary edge image, and filter processing is carried out; Hough transform is used for detecting straight lines of the binary edge image and a straight line cross point; and according to the distance between the camera and the moving target and the actual size of the target, the moving target is determined with the straight line cross point as the center. Thus, the operation speed is quick in the case of single moving target tracking based on the fixed shape under the dynamic background, the real-time performance for target tracking is ensured to be realized in low-speed movement, and the method is simple and easy to realize.

Description

technical field [0001] The invention relates to the technical field of target tracking, in particular to a method and device for tracking a single moving target based on a fixed shape in a dynamic background. Background technique [0002] In recent years, moving target tracking has become more and more widely used, especially with the rise of unmanned aerial vehicles (UAVs), its corresponding target tracking methods have been paid more and more attention. Tracking can be mainly divided into three categories: area-based tracking, contour-based tracking and feature-based tracking, and the premise of accurate tracking is to accurately detect the tracked target in the background. [0003] When detecting the target to be tracked, the first is the detection method of the target area. In the static background, the detection methods of moving target areas include frame difference method, optical flow method, background subtraction method, etc. Among them, the optical flow method ha...

AI Technical Summary

Problems solved by technology

For the straight line representation method in the geometric shape representation method, when performing target shape detection, the most commonly used method for detecting straight lines is the probability Hough transform, which can detect straight line segments, and delete all points on the straight line after finding the straight line segment , but the probabilistic Hough transform in the prior art needs to perform Hough space transformation again during the deletion process, which has a large amount of calculation and a relatively slow calculation speed. It is difficult to realize real-time processing on a platform with limited hardware conditions.

[0005] In addition, the Hough transform processes binary edge images. In the process of converting them into binary edge images, the edge detection algorithm canny algorithm will be used. However, the canny algorithm first needs to perform Gaussian filtering on the image, and the hardware conditions are limited. On the platform, the calculation speed cannot reach

[0006] In short, the single moving target tracking method based on a fixed form in the prior art has a large amount of calculation and a large amount of calculation, and it is difficult to achieve real-time tracking of the target

Images