Unlock instant, AI-driven research and patent intelligence for your innovation.

Moving target tracking method based on scale adaptive block particles

A scale-adaptive, target tracking technology, applied in the field of computer vision, can solve the problems of inability to fully express targets, large differences in tracking performance, and tracking failures

Pending Publication Date: 2019-06-07
KUNMING UNIV OF SCI & TECH
View PDF6 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] The technical problem to be solved by the present invention is to provide a scale-adaptive-based moving target tracking method to solve the defects that the existing targets cannot be fully expressed and the tracking performance varies greatly under different target scales and scenarios. , it is easy to add wrong information to the target model and cause the problem of tracking failure
[0007] The technical solution adopted in the present invention is: a moving target tracking method based on scale-adaptive block particles, which initializes the target area, and considers the overall information in the target frame when tracking the target in the current moving target method, but the shape of the tracked target has multiple Denaturation, which leads to the problem of low robustness and low accuracy of the tracking method

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 more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Moving target tracking method based on scale adaptive block particles
  • Moving target tracking method based on scale adaptive block particles
  • Moving target tracking method based on scale adaptive block particles

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0045] Embodiment 1: as figure 1 As shown, the moving target tracking method based on scale-adaptive block particles, the specific steps of the method are as follows:

[0046] Step 1. Initialize the target and select the target area; according to the first frame of the input image, with the target position as the center, collect an image block X whose size is twice the size of the target as the target area.

[0047] Step2. According to the Monte Carlo method, the target area, that is, the tracked target, is expressed as a group of block particles:

[0048] Step2.1. From Bayesian theory, it is assumed that the probability density function at time t-1 is known as: p(x t-1 |z 1:t-1 ), then when predicting, the probability density p(x t-1 |z 1:t-1 ) to get p(x t |z 1:t-1 ). Therefore we can take time t as z t medium particle x t The reliability probability density function is expressed as:

[0049] p(x t |z 1:t-1 )=∫p(x t |x 1:t-1 )p(x t-1 |z 1:t-1 ) ⑴

[0050] wh...

Embodiment 2

[0110] Embodiment 2: this method is compared with other methods in this embodiment, and its result is as follows Figure 2-4 And as shown in Table 1-2. figure 2 with image 3 Indicates the factor for the fast moving of the tracked target. When the target moves fast, this algorithm has a better effect than other algorithms; Figure 4 with Figure 5 Indicates the factors for the scale change of the tracked target. When the target changes from far to near or from near to far, the target scale will change. This algorithm has a better effect than other algorithms.

[0111] The present invention uses the average tracking error Center Location Error (CLE) and the overlap rate Pascal VOCOverlap Ratio (VOR) to evaluate this algorithm and other comparison algorithms. CLE refers to the deviation between the center of the object box and the center of the true location of the object. The smaller the center error, the higher the performance of the algorithm, and the larger the overlap ...

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

No PUM Login to View More

Abstract

The invention relates to a moving target tracking method based on scale adaptive block particles, and belongs to the field of computer vision. The method comprises the following steps: in a first frame of image, firstly initializing a target area, and taking a tracking target as a block particle by utilizing a local structure of the tracked target; under a sequence Monte Carlo method, constructinga block particle confidence function and a similarity function, and utilizing a scale pool method to select an optimal scale to obtain an optimal response position and an optimal response scale. According to the method, the tracked target can be described more comprehensively and specifically by using the localization method, and the local features of the target can be obtained while the global features of the target are obtained, so that the accuracy of target detection is improved. Secondly, the target scale is updated through a scale pool method; finally, the optimal target is selected byusing the peak sidelobe ratio of the response diagram, and the effectiveness of the model is improved.

Description

technical field [0001] The invention discloses a moving target tracking method based on scale self-adaptive block particles, which belongs to the field of computer vision. Background technique [0002] Object tracking is a hot spot in the field of computer vision, which is widely used in video surveillance, robot learning, industrial intelligence, etc. Its essence is to find the position and state of the target in a continuous video sequence image. Although object tracking has made great progress, it is still a challenging problem due to many factors such as occlusion, illumination change and scale change. [0003] Most current object tracking methods usually use bounding boxes to represent the extent of the tracked object. However, most of the tracked targets are not strictly rectangular in shape. Therefore, when the bounding box represents the tracked target, redundant background information is usually added to the model. Too much background information may reduce the ov...

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
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06T7/246
Inventor 尚振宏刘晓栋
Owner KUNMING UNIV OF SCI & TECH