Sea and air background weak target anti-interference tracking method based on random finite set

Abstract

The invention relates to a sea-sky background weak and small target anti-interference tracking method based on a random finite set. The method comprises the following steps: acquiring any ith frame of image containing a specified weak and small target in an image group; acquiring trace point positions of all weak and small targets of an (i + 1) th frame of image in the image group, and acquiring a multi-target track model of the (i + 1) th frame of image by using a random finite set method on the basis of a specified weak and small target of the ith frame of image; outputting the position of a specified weak and small target of the (i + 1) th frame of image according to each track obtained by the multi-target track model; according to the sea and air background weak and small target anti-interference tracking method based on the random finite set, in single target tracking application of infrared weak and small targets, the single target tracking problem is converted into multi-target detection, a track motion analysis method is combined, true and false targets are distinguished from a motion model, and the target tracking accuracy is improved. The problem that a weak target and a background interference point are difficult to distinguish is effectively solved; and a random finite set tracking technology is integrated, so that the tracking success rate is remarkably improved.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to an anti-interference tracking method for weak and small targets in sea and air background based on random finite sets. Background technique [0002] The single-target tracking technology for weak infrared targets is a major difficulty in photoelectric tracking technology. Considering that the appearance and disappearance of detected targets are random, and include real targets, interference targets, etc., the image characteristics of infrared weak and small targets are not obvious and change Multi-terminal, it is difficult to distinguish it from background interference points based on image data analysis. [0003] Traditional multi-target trajectory modeling methods use sensor measurements to match real targets. Inaccurate matching will lead to poor tracking performance. Only relying on weak target tracking based on image detection is extremely susceptible to background...

AI Technical Summary

Problems solved by technology

[0002] The single-target tracking technology for weak infrared targets is a major difficulty in photoelectric tracking technology. Considering that the appearance and disappearance of detected targets are random, and include real targets, interference targets, etc., the image characteristics of infrared weak and small targets are not obvious and change Multi-terminal, it is difficult to distinguish it from background interference points based on image data analysis

[0003] The traditional multi-target trajectory modeling method uses sensor measurements to match the real target. If the matching correlation is not accurate, the tracking effect will be deteriorated.

Only relying on weak target tracking based on image detection is extremely vulnerable to background interference, resulting in tracking failure

Images