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

Kernel correlation filtering algorithm for fast motion and motion blur

A correlative filter and motion blur technology, applied in character and pattern recognition, computing, computer components, etc., can solve problems affecting tracking accuracy and losing targets

Pending Publication Date: 2021-06-08
LIAONING TECHNICAL UNIVERSITY
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0007] The above methods will lose the target or affect the tracking accuracy when the target is moving fast and motion blurred.

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
  • Kernel correlation filtering algorithm for fast motion and motion blur
  • Kernel correlation filtering algorithm for fast motion and motion blur
  • Kernel correlation filtering algorithm for fast motion and motion blur

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0024] The specific implementation of the present invention will be described in detail below in conjunction with the accompanying drawings. As a part of this specification, the principles of the present invention will be described through examples. Other aspects, features and advantages of the present invention will become clear through the detailed description. In the referenced drawings, the same reference numerals are used for the same or similar components in different drawings.

[0025] KCF tracking algorithm: KCF (Kernelized Correlation Filter) target tracking algorithm is a discriminative tracking algorithm. The algorithm trains a target detector (correlation filter template) in the area where the target is located in the first frame to predict the position of the target in the next frame, and then updates the target detector with the new prediction result. When training the filter template, the target area is selected as a positive sample, and the target area is cycli...

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 coring correlation filtering algorithm for fast motion and motion blur. The method comprises the following steps: establishing an FM detector in a KCF algorithm, when the detector monitors that a target rapidly moves and a current frame tracking result is cancelled, expanding a current frame target searching range by taking a previous frame target frame as a center, and obtaining coordinate distribution of a possible position of the target for a related filter in the range by using a sliding window method; performing K-means clustering on the position distribution, finding out a distribution rule to obtain M position distribution domains, performing correlation filtering in the distribution domains by taking a class cluster center as a center to position a target center position, and obtaining M target center candidate points; and selecting an optimal target center point through a score indicator. According to the method, the robustness of the KCF algorithm to the fast motion and motion blur of the target is improved under the condition that the real-time performance is ensured, and a new thought is provided for multi-feature fusion.

Description

technical field [0001] The invention belongs to the technical field of graphics and image processing, and in particular relates to a kernelization correlation filtering algorithm for fast motion and motion blur. Background technique [0002] Visual object tracking is an important branch of computer vision, which has been widely used in video intelligent traffic monitoring, robotics, monitoring, human-computer interaction and other fields. According to the target appearance model, target tracking algorithms are divided into generative and discriminative tracking algorithms. In recent years, with the rise of machine learning, discriminative tracking methods have gradually occupied a dominant position because they can simultaneously use target information and background information around the target and have good tracking performance. [0003] Discriminative tracking methods learn a classification model to distinguish foreground from background through training data. The area...

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): G06K9/62G06K9/46
CPCG06V10/50G06F18/23213G06F18/2411
Inventor 崔丽群李明张俊东
Owner LIAONING TECHNICAL UNIVERSITY