Moving object tracking method under complicated background and sheltering condition

Abstract

The invention discloses a multi-feature and Kalman filtering combined object tracking method for realizing the effect of object tracking under a complicated background environment and a sheltering condition. Comprehensive feature based on color, texture and object moving information is provided for improving a CamShift algorithm, and a Kalman filter is combined to predict an object moving state, so that the tracking stability and the tracking accuracy of the moving target under a complicated background are improved. Color information is most likely to be interfered by factors such as illumination and background interference color; object texture feature is introduced to improve the defect when single color information is adopted; target moving information is continuously introduced; and further the interference in the background is removed. When an object is sheltered, least square fitting and target moving track extrapolation are performed through priori information before object shelter, the object moving positional information is predicted, and recapture of a moving object when the shelter is finished is facilitated.

Description

technical field [0001] The invention relates to an adaptive target tracking method, in particular to a moving target tracking method under complex background and occlusion conditions. Background technique [0002] The tracking of the moving target is to determine the position of the moving target of interest in each image of the video image, and correspond to the same target in different frames. The use of optical sensors to detect targets in complex backgrounds is of great significance to both civilian and military fields. The sensor design, information processing and system simulation methods involved in it have always been a hot research topic for researchers and engineers. [0003] In the process of detecting the surrounding environment by ground-based platforms (such as mobile and fixed), and in the process of ground-based detection by air-based platforms (such as aircraft, airships, satellites, etc.), the distance between the target and the background is often obtained...

AI Technical Summary

Problems solved by technology

But whether it is the Mean Shift algorithm or the CamShift algorithm, their essence is to use the color probability distribution information of the target, that is, the histogram is used as a feature during tracking, and the histogram is a weak feature, which leads to complex background or When there is a large area of ​​interference color in the background, it is easy to cause tracking failure

Moreover, the Mean Shift and CamShift algorithms do not make any predictions on the target's trajectory. When the target is fast maneuvering or occlusion is lost, it is often impossible to continue effective tracking.

[0007] In general, the current video tracking technology still has some difficulties in tracking moving targets, mainly from the following aspects: the impact of light intensity and weather changes on target tracking; changes in the appearance of targets or various complexities such as rotation, scaling, and displacement; the stability of tracking when the target is moving rapidly; the occlusion of the moving target; the influence of complex background interference and other factors; the correct detection and segmentation of the moving target; the problem of image data fusion, such as multi-camera tracking; real-time issues, etc.

[0008] Although the problem of video target tracking has been extensively studied and made great progress in the past ten years, many situations in the real environment will affect the reliable observation of the target in the video image, so the design of the system can be used in various complex environments. The method of accurately, quickly and stably tracking video targets is still a challenging task and an urgent problem to be solved, especially the moving target tracking method that can be applied to complex backgrounds and occlusions

Images