Multiple characteristic point tracking method for detecting shielded object

Abstract

The invention discloses a multiple characteristic point tracking method for detecting a shielded object. The method is characterized in that the moving object tracking method is based on characteristic optical flow segmentation and Kalman filtering estimation. According to the method, matching algorithms from coarse to fine levels are used to calculate characteristic optical flows, so that the matching precision is higher; an effective shielding decision algorithm is used, so that positions of characteristic points can be estimated when the characteristic points are shielded or lost; and the problem that a moving object is difficult to track when the object moves rapidly or is shielded is effectively solved.

Description

technical field [0001] The present invention is related to computer vision, image understanding and computer graphics. When the target is fully occluded or partially occluded, the detection and tracking of the target becomes particularly difficult. The present invention relates to a solution to moving targets under such complex conditions. The detection and segmentation method uses the established feature point optical flow clustering algorithm and the occlusion judgment algorithm to realize reliable detection of occluded targets. Background technique [0002] The automatic tracking technology of moving targets under complex background is an important subject of image processing, pattern recognition and computer graphics technology. It has broad application prospects in military, national defense and industrial fields. The tracking system has two basic operations, namely, target image segmentation and target tracking, but when the moving target performs large maneuvers and e...

AI Technical Summary

Problems solved by technology

[0002] The automatic tracking technology of moving targets in complex background is an important subject of image processing, pattern recognition and computer graphics technology, and has a wide range of applications in the fields of military, national defense and industry. In the application prospect, the tracking system has two basic operations, that is, the segmentation of the target image and the tracking of the target, but when the moving target performs large maneuvers and experiences occlusion, the stability, adaptability and computational efficiency of these operations will be affected. These effects will directly change the performance of the system, therefore, obtaining a robust tracking algorithm is an urgent problem to be solved.

[0003] Using motion estimation to segment and track moving objects is an important technical approach. The optical flow algorithm based on feature points and the algorithm based on continuous optical flow field are often used The motion estimation technology adopted, the characteristic optical flow is to obtain the optical flow at the feature point through feature matching. Compared with the continuous optical flow field algorithm, this algorithm has the ability to handle large inter-frame motion, is not sensitive to noise and calculation The characteristic of small quantity. But because of the sparse optical flow field, it is difficult to extract the precise shape of the moving target

Images