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

Scale self-adaptive kernel correlation filtering tracking method based on background perception

A technology of kernel correlation filtering and scale adaptation, applied in the field of target tracking, can solve problems such as hindering the performance improvement of trackers, difficult to deal with, and imperfect feature maps, and achieve the effect of accurate and rapid prediction of scale changes and improved robustness.

Active Publication Date: 2019-09-06
湖州云通科技有限公司
View PDF12 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, due to the complexity of visual tracking, the existing correlation filtering tracking algorithms are still difficult to deal with all situations in the tracking process, including target deformation, scale change, background similarity, etc.
The difficulties that hinder the performance improvement of the tracker mainly include the imperfection of the feature map, the boundary effect of the correlation filter, and the inability to accurately predict the change of the target scale in time, etc.

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
  • Scale self-adaptive kernel correlation filtering tracking method based on background perception
  • Scale self-adaptive kernel correlation filtering tracking method based on background perception
  • Scale self-adaptive kernel correlation filtering tracking method based on background perception

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0052] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some, not all, embodiments of the present invention.

[0053] see figure 1 , the present invention provides a technical solution: a background-aware scale adaptive kernel correlation filter tracking method, the method includes the following steps:

[0054] Step 1: Image preprocessing, performing histogram equalization processing on the image to reduce the influence of illumination changes.

[0055] Step 2: Obtain initial frame data and target information, including coordinates (x, y) of the upper left corner of the target area, target width w and height h.

[0056] Step 3: Extract the feature map of the target area, obtain the Fast Histogram of Oriented Gradient (FHOG) feature of the target area as the feature map of th...

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 discloses a scale self-adaptive kernel correlation filtering tracking method based on background perception, which relates to the technical field of target tracking and comprises the following steps: firstly, establishing a kernel correlation translation filter tracking model of background perception according to given target information; and then, adding an additional one-dimensional scale filter on the basis of predicting the translation position, so that the scale change of the target is quickly and accurately predicted. According to the scale self-adaptive kernel correlationfiltering tracking method based on background perception, a one-dimensional scale self-adaptive filter is added on the basis of a kernel correlation filter based on background perception, and the position and scale change of a target can be accurately and quickly predicted on the basis of solving the boundary effect of the correlation filter; the method has good adaptability in a complex trackingscene.

Description

technical field [0001] The invention relates to a target tracking technology, in particular to a background perception-based scale adaptive kernel correlation filter tracking method. Background technique [0002] Due to the rapid development of computer hardware in recent years, relevant algorithms have been continuously proposed. Correlation filtering visual tracking algorithms have achieved a large number of excellent results and are widely used in video detection, intelligent driving, intelligent tracking and other fields. However, due to the complexity of visual tracking, the existing correlation filtering tracking algorithms are still difficult to deal with all situations in the tracking process, including target deformation, scale change, background similarity, etc. The difficulties that hinder the performance improvement of the tracker mainly include the imperfection of the feature map, the boundary effect of the correlation filter, and the inability to accurately pre...

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
IPC IPC(8): G06T7/246G06T5/40
CPCG06T7/246G06T5/40G06T2207/20004
Inventor 王永雄冯汉魏国亮张孙杰
Owner 湖州云通科技有限公司