Improved TLD Tracking Method Based on Improved Online Boosting and Kalman Filter

Abstract

The invention proposes an improved TLD tracking method based on improved Online Boosting and a Kalman filter, belonging to the technical fields of machine vision, artificial intelligence, human-computer interaction, and target tracking. The method includes the following steps: (1) initialization: use the initial sample set formed by selecting the target and affine transformation to initialize the improved Online Boosting classifier and P-N learner; (2) image tracking: select feature points, use twice The L-K optical flow method is used to track it, and the tracking error is compared with the threshold twice to obtain the tracking result; (3) Image detection: after Kalman filter, variance classifier, Online Boosting classifier, KNN classifier (4) Integrated tracking and detection results: evaluate the confidence of the tracker and detector results, and obtain the final module result; (5) Online learning: P-N learner is used to correct the tracker, Detector results, and enrich the sample set. The invention can effectively overcome the occlusion problem, increase the speed of the original method, and simultaneously effectively improve the accuracy and robustness of the detector.

Description

technical field [0001] The invention relates to a TLD (Tracking-Learning-Detection tracking-learning-detection) tracking method based on improved Online Boosting (online cascade classifier) ​​and Kalman filter improvement, belonging to machine vision, artificial intelligence, human-computer interaction and Object tracking technology field. Background technique [0002] Video image tracking has always been the focus of attention in the field of computer and image. The early video tracking mainly used the target tracking technology based on feature matching, which mainly used the light and shade, edge, color, texture and temporal and spatial differences of the moving target in the image sequence to detect the moving object. Among them, the literature (Meanshift proposed by Comaniciu D, Meer P: a robust method for feature space analysis, published in the direction of IEEE pattern recognition and machine intelligence) and the literature (Allen J G, Xu R Y D, Jin J S proposed th...

AI Technical Summary

Problems solved by technology

However, when the target is occluded by a large area, the accuracy of the target tracking is severely reduced by this method, and when the initial tracking is performed, there are fewer samples, and the robustness of the detector is poor.

Images