Ant colony neighbor cell tracking system based on cooperation and competition mode and application thereof

An ant colony and cell technology, applied in the field of ant colony neighbor cell tracking system, can solve the problems of unresearched tracking performance, insufficient tracking accuracy and stability, etc.

Active Publication Date: 2014-07-30
HUAWEI TEHCHNOLOGIES CO LTD
View PDF4 Cites 10 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] Correctly tracking neighboring multi-cells is still a difficult problem in cell tracking. The current technology is rarely able to deal with these situations. The tracking ac

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
  • Ant colony neighbor cell tracking system based on cooperation and competition mode and application thereof
  • Ant colony neighbor cell tracking system based on cooperation and competition mode and application thereof
  • Ant colony neighbor cell tracking system based on cooperation and competition mode and application thereof

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0061] The initial ant colony distribution and rough classification module, the specific steps are:

[0062] 1) Cell image background extraction: use the approximate average method (Approximate Median Method) to extract the background image of the cell. The approximate average method is an iterative method that compares the pixel values ​​on the background image with the pixel values ​​of the current image. If the background If the pixel is large, the background pixel will be reduced by 1, otherwise it will be increased by 1, and the final background model will converge to a value. In this way, the background image can be obtained;

[0063] 2) Acquisition of the foreground of the cell image: compare the current frame image pixel with the corresponding pixel of the background model image, if the difference between the two is greater than a certain threshold, then this pixel belongs to the foreground, the formula is:

[0064]

[0065]

[0066] Where I(i) is the current fra...

Embodiment 2

[0079] An example of the present invention being used to analyze white blood cells, after the initial ant colony distribution and classification module, the multi-task ant colony module based on cooperation and competition mode, and the fusion and deletion module based on the integrated pheromone field, the neighbor multi-cell tracking results are as follows.

[0080] figure 2 Taking the 23rd frame in Sequence 2 as an example to illustrate the results of background extraction and initial ant colony distribution. image 3 For the classification results based on it, it can be seen that there are three cells in area A but they are divided into four types, which can be solved by our proposed multi-ant colony system. Figure 6-7 For the analysis of the sequence tracking results of sequence 1, it mainly tracks the adjacent multiple cells caused by movement, and the dynamic characteristics of the cells are different and the number of cells changes. The parameters are set to (ρ=0.7, ...

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 an ant colony neighbor cell tracking system based on a cooperation and competition mode and application thereof. The system includes an initial ant colony distribution and rough classification module, a multi-ant-colony decision module and a fusion and deletion module. The initial ant colony distribution and rough classification module uses an approximate median method to extract a foreground image of cells so as to obtain initial ant colony distribution and then uses a K-mean-value clustering method to roughly divides the ant colony into N groups of sub-ant colonies. The multi-ant-colony decision module constructs a cell position estimation module, which is under the common action of N independent sub-pheromone fields and total pheromone fields, on the cooperation and competition mode. The fusion and deletion module performs multi-cell position estimation based on pheromone field construction through combination of similar sub-ant-colonies and removal of false targets caused by clutter and uses a nearest neighbor method which is easy to realize to perform cell correlation and obtain cell motion trails and kinetic parameters. The ant colony neighbor cell tracking system based on the cooperation and competition mode is capable of realizing neighbor multi-cell kinetic parameter estimation under a low-contrast-ratio cell image sequence, that is, under conditions that neighbor multi-cell kinetic characteristics are different, cells are deformed and time variance happens in cell number and the like, on the basis that no cell detection module or large quantity of cell training samples are needed, through cooperation and competition of an ant cluster system, problems of neighbor multi-cell multi-parameter estimation and tracking are solved.

Description

technical field [0001] The invention relates to an ant colony neighbor cell tracking system based on cooperation and competition mode and its application. Background technique [0002] Cell movement analysis has very important application value in the field of cell biology. Cell tracking technology is one of the most effective tools for studying cell dynamics, and it is a very challenging and highly applicable topic in both information science and life science. Cell movement analysis faces many challenges, such as cell neighbors, cell division, differences in dynamic properties of multiple cells, cell overlap, cell number changes, low-contrast ratio cell image sequences, etc. The traditional manual method is a tedious and time-consuming process with less data to process. The invention intends to use the intelligent search behavior of the ant colony to study the movement characteristics of the adjacent multi-cells, and to design the corresponding automatic tracking technolo...

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): G06T7/20G06N3/00
Inventor 鲁明丽徐本连朱培逸毛丽民施健任亚运
Owner HUAWEI TEHCHNOLOGIES CO LTD
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap