A Self-Recovery Target Tracking Method Based on Multi-level Detector
A target tracking and detector technology, applied in instruments, image analysis, image enhancement, etc., can solve the problems of decreased detection target accuracy, affecting tracking effect, tracking error, etc., to achieve high engineering application value, improve tracking performance, and improve accuracy. degree of effect
Active Publication Date: 2018-12-18
BEIHANG UNIV
View PDF3 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
However, in the actual process of using detection to track the target, the common problem of the above algorithms is the tracking error caused by various possible factors. During the tracking process, the error gradually accumulates, which will lead to the decline in the accuracy of the subsequent detection target, and finally Affect Tracking Effect
In addition, in the initialization stage of target tracking, the acquired target initial template itself has accuracy problems, which will seriously affect the accuracy of the subsequent tracking process.
If there are obstacles or buildings in the trajectory of the target's movement, the target's features will disappear partially or even completely due to being blocked. When the target reappears, it will not be able to re-detect and continue tracking
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0065] The present invention will be described in detail below in conjunction with the accompanying drawings and embodiments.
[0066] Such as figure 1 Shown, the concrete realization steps of the present invention are as follows:
[0067] (1) Build a saliency detector
[0068] After initializing the first frame of image, determine the location of the target and the size of the area, and superimpose appropriate Gaussian noise and rotation transformation on the image block where the target is located, generate a series of positive sample sets of the target, and then randomly collect several such samples in the area outside the target. Large background image patches, as a set of negative samples. When the following image frames are input, the salient regions in the image are first detected by a saliency detector.
[0069] The principle of the saliency detector is to calculate the Euclidean distance between each pixel and the average brightness value of all pixels in the entir...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
A self-recovery target tracking method based on multi-level detectors: combined with the idea of cascading Adaboost multi-level weak classifiers into strong classifiers, multiple different types of detectors are selected in series; the first-level detection selects the salience Detector; the second stage selects the classifier set detection module, and substitutes it into the random tree to calculate the posterior probability that the possible area detected by the first stage is a positive sample; the third stage selects the correlation filter detector, and calculates the posterior probability greater than a certain The correlation between the sample of the threshold value and the positive sample obtained in the initialization or the previous frame, thereby reducing the accumulated error caused by long-term tracking; after the multi-level detector is determined, the position with the highest correlation is identified as the target area in the current frame, and Sampling at a certain location supplements the number of positive and negative samples removed in the sample set to ensure the reliability and number of samples are constant; if the maximum correlation degree is less than a certain threshold, the re-detection mechanism is activated to re-detect the search target in the area near the location.
Description
technical field [0001] The present invention designs a target tracking method with self-recovery capability based on multi-level detectors, and connects multiple detectors of different types in series to form a strong detector to detect the target on each frame of the video sequence, thereby realizing the target Tracking, and can perform re-detection when the tracking target is lost, re-detect the target position and continue tracking. This method adopts the idea of target tracking based on detection, which has good robustness to static background and dynamic background, and has high engineering application value. Background technique [0002] Theoretical research on how to maintain long-term stable tracking of image targets has gradually become a direction of higher concern in computer vision, which has been widely used in monitoring and recognition and robotics and other fields. Object tracking is a comprehensive research and development direction that integrates many d...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & AuthorityPatents(China)
IPC IPC(8): G06K9/46G06K9/62G06T7/136G06T7/246G06T7/262
CPCG06T7/136G06T7/246G06T7/262G06T2207/20056G06V10/443G06V10/462G06V2201/07G06F18/2415
Inventor张弘饶波李伟鹏
OwnerBEIHANG UNIV