Biological image denoising algorithm based on statistical rule

An image and biological technology, which is applied in the field of optical microscopy and metrology microscopic imaging, can solve the problems of not meeting engineering needs and not being able to completely remove noise, and achieve the effect of improving the removal range and removal accuracy

Active Publication Date: 2017-02-15
HARBIN INST OF TECH
View PDF6 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to solve the problem that the traditional algorithm processing results cannot completely remove the noise, which does not meet the

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
  • Biological image denoising algorithm based on statistical rule
  • Biological image denoising algorithm based on statistical rule
  • Biological image denoising algorithm based on statistical rule

Examples

Experimental program
Comparison scheme
Effect test

Example Embodiment

[0020] Specific implementation manner 1: The biological image denoising algorithm based on statistical laws in this embodiment includes three steps: selecting filter templates, cell edge and noise monitoring, and image denoising processing. The specific process is:

[0021] Step 1. Select a filter template: use a biological microscope to collect RGB images of biological cells, select a filter template that can separate the cells from isolated noise from the RGB image collected by the traversal biological microscope, and the selected filter templates are the mean filter template and Gaussian Filter template, RGB image is composed of a single cell and stray noise in the background area, or a cell group composed of multiple cells and stray noise in the background area;

[0022] Step 2. Cell edge and noise monitoring: Use the mean filter template and Gaussian filter template to traverse the entire collected biological cell image to determine the location of the biological cell edge and ...

Example Embodiment

[0026] Specific embodiment 2: This embodiment further explains the first embodiment. The specific process of using the mean filter template and the Gaussian filter template to traverse the entire biological cell image collected in step 2 to determine the edge position of the biological cell and the position of the isolated noise point is :

[0027] Use the mean filter template and the Gaussian filter template to traverse the entire collected biological cell image, use the Gaussian filter template in each area to obtain the filter threshold of the region, and then use the mean filter template to select the region greater than or based on the determined filter threshold. If the number is less than the filtering threshold, it is determined according to the number whether the traversed area is the isolated point noise position or the cell edge or internal position.

Example Embodiment

[0028] Specific embodiment three: This embodiment further explains the second embodiment. It is assumed that the brightness value threshold obtained by the Gaussian filter template is S=48. When the brightness value in the area traversed by the mean filter template is greater than S=48, the number exceeds the mean filter When the traversed area is located at the edge or inside of the cell when the template is 1 / 2, when the brightness value inside the area traversed by the mean filter template is greater than S=48 and does not exceed 1 / 2 of the mean filter template, this The traversed area is located at the outlier noise position.

[0029] In this embodiment, the location of the original image is selected to be filtered or retained according to the determined area division of the image, so as to achieve the effect of accurate filtering. The image effect after filtering according to this method is as follows: Figure 4 Shown.

[0030] In this embodiment, according to figure 2 with i...

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

A biological image denoising algorithm based on a statistical rule belongs to the optical microscopic technology and metering microscopic imaging technology fields. By using a traditional algorithm, in a processing result, a noise can not be completely removed so that an engineering need can not be satisfied, and many limitations exists during image processing. The algorithm comprises three steps of filtering template selection, cell edge and noise monitoring, and image denoising processing. A specific process comprises the following steps of using a biological microscope to collect RGB images of biological cells and selecting filtering templates to be a mean value filtering template and a Gaussian filtering template; using the mean value filtering template and the Gaussian filtering template to traverse a collected whole biological cell image and determining a biological cell edge position and an isolated noise point position; and carrying out filtering processing on the biological cell edge position and the isolated noise point position, filtering a isolated point noise, retaining a biological cell sample graph and acquiring a denoised biological cell image. The algorithm is used for biological image denoising.

Description

technical field [0001] The invention relates to a biological image denoising algorithm, which belongs to the technical field of optical microscopic technology and measurement microscopic imaging. Background technique [0002] On the basis of the three-dimensional height information of the sample measured by the confocal microscope, one-dimensional color information is added to form a four-dimensional information map that can directly reflect the material and shape of the sample. The three-dimensional height information can be determined by the tomographic data measured by the confocal microscope. The processing results are relatively accurate, and the color information of each position of the sample is obtained by taking a photo of a color CCD, but the brightness and contrast of the picture obtained by taking a color CCD are not high, and contain many isolated noise points, resulting in the possible distortion of the color picture. The recognition is not good. When we use th...

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): G06T5/00
CPCG06T5/002G06T2207/20024
Inventor 刘俭谭久彬李勇牛斌
Owner HARBIN INST OF TECH
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