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

Image segmentation method based on feature extraction and denoising

A feature extraction and image segmentation technology, applied in the field of image processing, can solve the problems of inaccurate segmentation, interference with gestational sac segmentation, and the inability of algorithms to be independently competent.

Active Publication Date: 2019-07-26
HUNAN ZIXING INTELLIGENT MEDICAL TECH CO LTD
View PDF9 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The classic algorithms involved in the current image segmentation method are already quite mature, but none of the algorithms can be independently competent in solving practical problems.
Among the segmentation algorithms, the threshold segmentation and active contour segmentation technology segmentation algorithms are quite mature, but the segmentation of gestational sac and yolk sac-germ in this study will be affected by effusion, noise, etc., resulting in inaccurate segmentation
The difficulty in the segmentation of the gestational sac is that there are various effusions in the B-ultrasound image, and some of the effusions resemble the gestational sac, so it interferes with the segmentation of the gestational sac; and the segmentation of the yolk sac-germ is due to the fact that most of the yolk sac - The germs are all on the edge of the gestational sac and the gray features of the edge are very similar to those outside the gestational sac, so it is not easy to segment. The existing image processing methods cannot effectively analyze the gestational sac, yolk sac-germ of the B-ultrasound image effective processing

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
  • Image segmentation method based on feature extraction and denoising
  • Image segmentation method based on feature extraction and denoising
  • Image segmentation method based on feature extraction and denoising

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0088] An image segmentation method based on feature extraction and denoising, including the gestational sac segmentation process and the yolk sac-germ development state segmentation process;

[0089] The gestational sac segmentation process includes the following steps:

[0090] Step 1: Input B-ultrasound images such as figure 1 (a) and figure 2 (a);

[0091] Step 2: Obtain the binary image of the B-ultrasound image through image segmentation and feature extraction;

[0092] Step 3: Select the area where the gestational sac is located to obtain the binary image of the gestational sac, such as figure 1 (b) and figure 2 (b), the processing steps are as follows:

[0093] 3.1) Use the mask method to select the area where the gestational sac is located to obtain an image;

[0094] 3.2) According to the image binary image obtained in 3.1, remove the edge noise to obtain the edge noise-removed image.

[0095] Step 4: Image feature extraction, the processing steps are as fol...

Embodiment 2

[0143] Such as figure 1 Shown is the division of the gestational sac and yolk sac

[0144] (1) Image description;

[0145] (2) Read in an original B-ultrasound grayscale image src_before, as attached figure 1 (a);

[0146] (3) Use the level set algorithm to segment the image of src_before, and then select the area where the gestational sac is located according to the segmented B-ultrasound image (in order to remove the effusion, dark areas, etc., and the grayscale features are similar to the gestational sac part to segment the gestational sac The impact of the results), and then perform feature extraction and denoising operations by observing the characteristics of the gestational sac, and finally obtain the binary image GS_img of the gestational sac, as shown in the attached figure 1 (b);

[0147] (4) Grayscale the binary image GS_img of the gestational sac to obtain the binary result image GS_Gray of the gestational sac and output it, as attached figure 1 (c);

[0148]...

Embodiment 3

[0154] Such as figure 2 Shown is the division of the gestational sac and germ

[0155] (1) Image description;

[0156] (2) Read in an original B-ultrasound grayscale image src_before, as attached figure 2 (a);

[0157] (3) Use the level set algorithm to segment the image of src_before, and then select the area where the gestational sac is located according to the segmented B-ultrasound image (in order to remove the effusion, dark areas, etc., and the grayscale features are similar to the gestational sac part to segment the gestational sac The impact of the results), and then perform feature extraction and denoising operations by observing the characteristics of the gestational sac, and finally obtain the binary image GS_img of the gestational sac, as shown in the attached figure 2 (b);

[0158] (4) Grayscale the binary image GS_img of the gestational sac to obtain the binary result image GS_Gray of the gestational sac and output it, as attached figure 2 (c);

[0159]...

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 image segmentation method based on feature extraction and denoising, and the method comprises the steps: firstly carrying out the segmentation to obtain a B ultrasonic image, and obtaining a pregnancy sac through the selection of an area where the pregnancy sac is located, feature extraction, denoising, convex hull detection and other algorithms; carrying out AND operation on a level set segmentation result graph and the gestational sac, and segmenting out the egg yolk sac-germ developmental state through the algorithms of feature extraction, morphology, denoising and the like; In the selected 149 pieces of data, the pregnancy sac segmentation accuracy rate is 94%, and the yolk sac-germ segmentation segmentation accuracy rate reaches 77.14%.

Description

technical field [0001] The invention relates to a method for segmenting a gestational sac and a yolk sac-germ development state from a B-ultrasound image, and belongs to the field of image processing. Background technique [0002] Prenatal big data can monitor the whole process of embryonic development, automatically identify and measure key developmental indicators in each stage of embryonic growth, such as gestational sac, yolk sac-embryo. After our machine deeply learns the doctor's knowledge and experience, an embryonic development evaluation system can be established to automatically judge whether the embryonic development is normal, so as to intervene early. [0003] The classic algorithms involved in the current image segmentation methods are quite mature, but none of the algorithms can be independently competent in solving practical problems. Among the segmentation algorithms, threshold segmentation and active contour segmentation technology segmentation algorithms ...

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): G06T5/00G06T7/136G06T7/155G06T7/187
CPCG06T7/136G06T7/155G06T7/187G06T2207/10132G06T2207/30044G06T5/70Y02A90/10
Inventor 石慧娟刘丽珏李仪穆阳
Owner HUNAN ZIXING INTELLIGENT MEDICAL TECH CO LTD
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