Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

A pedestrian target tracking method based on multi-example learning

A multi-instance learning, pedestrian target technology, applied in the direction of instruments, character and pattern recognition, computer components, etc., can solve the problems of large changes in target appearance, easy tracking failure, etc., to reduce the possibility and accuracy of tracking failure Improved effect

Active Publication Date: 2019-06-21
SHENZHEN POLYTECHNIC
View PDF4 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to provide a pedestrian target tracking method based on multi-instance learning to solve the problem that the target model established according to the first frame is prone to tracking failure when the target appearance changes greatly.

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
  • A pedestrian target tracking method based on multi-example learning
  • A pedestrian target tracking method based on multi-example learning
  • A pedestrian target tracking method based on multi-example learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0039] 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.

[0040] see Figure 1-2 , an embodiment provided by the present invention: a pedestrian target tracking method based on multi-instance learning, comprising the following steps:

[0041] S1, intercept image data;

[0042] S2. Decompose the image data into several block graphics, the decomposition depends on the resolution of the image and the relevant pixel level, and can accurately locate and track the target;

[0043] S3, extracting the graphic features of the block graphics, including extracting facial features, action features, color features and shape features;

[0044] S4, comparing the graphic features of the block graphics and the features of pedestrian targe...

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 pedestrian target tracking method based on multi-instance learning, relates to the related field of target tracking methods, and aims to solve the problem that a target modelestablished according to a first frame in the prior art is easy to fail in tracking when the appearance of a target is greatly changed. The method comprises the following steps: S1, intercepting image data; S2, decomposing the image data into a plurality of block graphs; S3, extracting graphic features of the block graph; S4, comparing the graphic features of the block graph with the features ofthe pedestrian target; S5, constructing a classifier; S6, calculating the overlap ratio score of the examples in each classification pool of the classifier; S7, performing weight calculation on the coincidence degree score; S8, calculating and tracking a pedestrian target according to the weight, and the comparison method in the step S4 comprises the following steps: S1, stripping color features of a block graph; S2, establishing a color coincidence degree classification pool; S3, stripping the shape characteristics of the block graphs; And S4, establishing a shape coincidence degree classification pool.

Description

technical field [0001] The invention relates to the related field of target tracking methods, in particular to a pedestrian target tracking method based on multi-instance learning. Background technique [0002] In machine learning, multi-instance learning is evolved from supervised learning. Compared with inputting a series of individually labeled examples, in multi-instance learning, the input is a series of labeled "packages", each " A package” includes many examples. When all examples in the package are negative examples, the package will be labeled as a negative package; when the package contains at least one positive example, the package will be labeled as a positive package. When receiving a series of labeled packets, the machine tries to: (1) generalize a class concept in order to correctly label individual examples; (2) learn how to label a packet in addition to generalization. [0003] The commonly used target model is a static model. In the initial stage of tracki...

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): G06K9/00G06K9/46G06K9/62
Inventor 连国云孙宏伟
Owner SHENZHEN POLYTECHNIC
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com