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

Full-slice digital imaging self-adaptive automatic focusing method based on transfer learning

A technology of automatic focusing and transfer learning, which is applied in graphics and image conversion, image data processing, instruments, etc., to achieve the effect of ensuring the accuracy of automatic focusing, improving generalization ability and adaptability

Active Publication Date: 2020-12-11
HARBIN INST OF TECH
View PDF2 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] Considering the limitations of traditional methods, the present invention uses advanced machine learning algorithms to solve the automatic focus problem of full-slice digital imaging

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
  • Full-slice digital imaging self-adaptive automatic focusing method based on transfer learning
  • Full-slice digital imaging self-adaptive automatic focusing method based on transfer learning
  • Full-slice digital imaging self-adaptive automatic focusing method based on transfer learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0024] The specific implementation of the method of the present invention will be further described below.

[0025] combine Figure 1 to Figure 3 As shown, a method for self-adaptive automatic focusing of full-slice digital imaging based on migration learning disclosed in this embodiment includes the following steps:

[0026] Step a, input out-of-focus image;

[0027] Step b, automatically focusing on the network;

[0028] Step c, predicting the quasi-focus distance;

[0029] Step d, new data training,

[0030] Using the method of transfer learning, the adaptive automatic focusing method under different data is realized.

[0031] Specifically, the input defocused images come from z-stack image stacks obtained by axial scanning movement at different sub-image lateral positions, and each sub-image position obtains 20 positive and negative defocused images and one in-focus image , a total of 41 pieces.

[0032] Specifically, the autofocus network is pre-trained according to...

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 belongs to the field of biomedical instruments, and particularly relates to a full-slice digital imaging self-adaptive automatic focusing method based on transfer learning. The transferlearning method based on a neural network mainly comprises an automatic focusing network, wherein the designed automatic focusing network can calculate the focusing distance according to the out-of-focus image; gradually adding the data under the new data set in an iterative mode to carry out transfer learning training; by adopting a transfer learning training mode, the automatic focusing networkcan have generalization and adaptivity capable of meeting requirements, data acquisition of different biological cell samples can be realized, and software virtualization is carried out on traditionalfull-slice digital pathological imaging hardware.

Description

technical field [0001] The present invention is based on the field of biomedical instruments, takes deep learning technology as the core, and specifically relates to a full-slice digital imaging adaptive automatic focusing method based on transfer learning, which can be widely used in the fields of instrument science, artificial intelligence, medical imaging and automation, etc. Research. Background technique [0002] In recent years, advanced digital pathology imaging technology has been widely researched and applied. Whole Slide Images (WSI, Whole Slide Images), that is, virtual microscopy, can collect traditional microscopic sections in the form of digital images, which can achieve arbitrary computer access, easy storage, and communication between researchers and doctors. Off-site transmission, etc. Whole slide digital imaging is crucial in bioimaging research, such as in areas such as cancer analysis and disease prediction. At present, the US Food and Drug Administrat...

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): G06T3/00G06N3/04G06F30/27
CPCG06F30/27G06N3/045G06T3/04
Inventor 刘贤明李强
Owner HARBIN INST OF TECH
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