A target tracking method based on target and camera motion modeling

Abstract

The invention discloses a target tracking method based on target and camera motion modeling. The target tracking method comprises the following steps: S1, predicting the real speed of a target at thecurrent moment through a filter algorithm; S2, predicting the speed of the camera at the current moment; S3, synthesizing the real speed of the target predicted in the step S1 and the speed of the camera predicted in the step S2 to obtain a predicted value of the resultant velocity of the target in the pixel plane at the current moment, and selecting a target search area according to the predictedvalue of the resultant velocity of the target in the pixel plane; and S4, carrying out a target tracking algorithm in the target search area selected in the step S3, and obtaining the position of thetarget in the pixel plane at the current moment. According to the target tracking method based on target and camera motion modeling provided by the invention, the accuracy and robustness of target tracking are effectively improved.

Description

technical field [0001] The invention relates to the field of computer vision, in particular to an object tracking method based on object and camera motion modeling. Background technique [0002] Video analysis has a wide range of application scenarios in real life. The main steps of video analysis include object detection, object tracking, and behavior analysis. Target detection is to find the target we are interested in, and target tracking can maintain continuous tracking of the target and evaluate the state information of the target in each video frame. Behavior analysis is based on the continuous tracking information of the target state, discovering the target behavior characteristics, and Provide a basis for possible decisions. From the analysis, it can be found that object tracking plays an extremely important role in video analysis and has a strong practical demand. [0003] Visual target tracking can be divided into preprocessing, feature extraction, target state ...

AI Technical Summary

Problems solved by technology

The few methods that use motion features are also due to the fact that motion features perform poorly in many cases

[0004] The reason is that after the target is imaged by the camera, due to perspective and other reasons, the constraints on the target movement in the real scene may be greatly damaged. What is more serious is that the target tracking scene is not only used for camera fixation. Motionless scenes are also widely used when the camera is moving, so the movement of the camera can directly cause the target to move irregularly on the image plane, that is, the constraints of the physical world are broken, and it can be used for target tracking. The prior knowledge is lost, so those methods that directly model the motion of the image plane target get no longer real and reliable motion information, making the tracking effect very poor

Images