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

Multi-target model visual tracking method based on cost-sensitive three-way decision

A multi-objective model, cost-sensitive technology, applied in the field of surveillance video intelligent analysis, can solve the problems of lack of multiple features, multiple decision types, poor accuracy and robustness, etc., to improve accuracy, good adaptability and stability. The effect of robustness

Active Publication Date: 2020-06-05
TONGJI UNIV
View PDF10 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Although the addition of spectral information helps to improve the tracking effect, the traditional method does not have the advantages of multi-features and multi-decision types, and its accuracy and robustness are not good.

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
  • Multi-target model visual tracking method based on cost-sensitive three-way decision
  • Multi-target model visual tracking method based on cost-sensitive three-way decision
  • Multi-target model visual tracking method based on cost-sensitive three-way decision

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0048] In order to make the object, technical scheme and advantages of the present invention clearer, below in conjunction with embodiment, specifically as figure 1 The shown algorithm flow chart further describes the present invention in detail. It should be understood that the specific embodiments described here are only used to explain the present invention, but not to limit the present invention.

[0049] Step 1: Establish multiple target models with different update strategies. The specific description is as follows: The target model should have the following characteristics: keep the characteristics of the tracked target unchanged, adapt to the appearance changes of the target in time, and be able to predict the future appearance changes of the target. Therefore, the present invention firstly saves the target appearance of the initial frame as a constant target feature; secondly, the target appearance updated frame by frame is used as a target feature that can adapt to c...

specific Embodiment approach

[0067] figure 1 It is an algorithm implementation flow chart of the present invention, and the specific implementation is as follows:

[0068] 1. Establish multiple target models with different update strategies to form a set

[0069] 2. Use similarity learning model tracking to obtain multi-model candidate frames s ij , used in the calculations of subsequent Articles 3 and 6;

[0070] 3. Calculate the number N of overlapping pairs of all candidate frames t ;

[0071] 4. Set the cost dictionary for each decision type, according to N t Calculate decision cost;

[0072] 5. Based on the cost-sensitive three-way decision-making theory, calculate the upper and lower boundaries α and β of the three-way decision-making according to the decision cost;

[0073] 6. According to the upper and lower boundaries α and β, use the convolutional neural network discriminator to classify the candidate frames obtained in the second step;

[0074] 7. Use the candidate frame with the highe...

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 multi-target model visual tracking method based on a cost-sensitive three-way decision. The method comprises the following steps: 1) establishing a multi-target model Zi inan ith frame; 2) searching a multi-target set xi + 1 in the (i + 1) th frame by using the multi-target model Zi; 3) according to the position overlapping condition of the samples in the multi-target set xi + 1, respectively solving cost values lambda of three decision results of positive-acceptance, positive-rejection, negative-acceptance, negative-rejection, intermediate-acceptance and intermediate-rejection; 4) solving a decision boundary based on a cost-sensitive three-way decision method, and dividing the multi-target set xi + 1 into a positive sample, a negative sample and an intermediatesample; 5) selecting a result s (i + 1) 1 with the highest confidence coefficient as a temporary tracking result, and recording the tracking result for later use; and 6) in the (i + 2) th frame, if s(i + 1) j exists in the (i + 2) th frame, providing a more reliable tracking result, replacing the tracking result provided by s (i + 1) 1 with the more reliable tracking result. Compared with the prior art, the method has the advantages of quickness, accuracy, good robustness and the like.

Description

technical field [0001] The invention relates to the field of monitoring video intelligent analysis, in particular to a multi-target model visual tracking method based on cost-sensitive three-way decision-making. Background technique [0002] Visual tracking refers to in a short video (generally less than 1000 frames, and there is no camera switching), for any type of determined target, after its position and size are given in the first frame, its position and size are monitored online in subsequent frames process. Compared with the target detection algorithm, this algorithm generally has the characteristics of fast speed and high accuracy, and is mainly used in the following scenarios: (1) Scenarios that need to identify the target motion trajectory; (2) Candidate frame labeling for auxiliary target detection; (3) ) cooperates with the target recognition algorithm to extract candidate targets, save computing power, and realize efficient and accurate recognition. Therefore,...

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): G06K9/00G06K9/62
CPCG06V20/41G06V2201/07G06F18/241
Inventor 赵才荣孙添力
Owner TONGJI UNIV
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