An Image Retrieval Method Based on Semantic Features and Supervised Hashing

A technology of semantic features and image retrieval, applied in image enhancement, image analysis, image data processing and other directions, can solve problems such as misdiagnosis and missed diagnosis

Active Publication Date: 2020-10-30
TAIYUAN UNIV OF TECH
View PDF8 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, doctors mainly diagnose diseases based on experience, and the diagnosis results are subjective to a certain extent, and misdiagnosis and missed diagnosis often occur.

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
  • An Image Retrieval Method Based on Semantic Features and Supervised Hashing
  • An Image Retrieval Method Based on Semantic Features and Supervised Hashing
  • An Image Retrieval Method Based on Semantic Features and Supervised Hashing

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0058] The present invention will be described in detail below in conjunction with specific embodiments.

[0059] refer to figure 1 , 2 , 3, 5, the realization process of the inventive method is as follows:

[0060] A method of image retrieval based on semantic features and supervised hashing to realize medical sign recognition of pulmonary nodules, comprising the following steps:

[0061] Step A, extracting the mixed sign area of ​​pulmonary nodules in the lung CT image, and intercepting each single sign area, extracting the semantic features expressing the sign information of the pulmonary nodule and retrieving similar pulmonary nodule images, and then identifying the query image Prepare for the medical signs presented;

[0062] Step B, using a convolutional neural network (CNN) based on parameter sharing to extract semantic features expressing pulmonary nodule sign information; first use the first CNN to train single sign data, and adjust network parameters to effectivel...

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 a method for realizing pulmonary nodule medical sign recognition based on semantic feature and supervised hash image retrieval, comprising the following steps: step A, extracting the mixed sign area of ​​pulmonary nodule in a lung CT image, and intercepting Each single sign area; step B, using convolutional neural network (CNN) based on parameter sharing to extract semantic features expressing pulmonary nodule sign information; step C, for realizing similar pulmonary nodule image retrieval; step D, using for identifying signs of pulmonary nodules. The method of the present invention is based on semantic features and supervised hashing of pulmonary nodule image retrieval, and then identifies the sign category represented by the pulmonary nodule image, which is convenient for doctors to judge the degree of benign and malignant pulmonary nodules, and reduces the doctor's excessive diagnostic experience. rely.

Description

technical field [0001] The invention relates to pulmonary nodule sign recognition, in particular to an image retrieval method based on semantic features and supervised hashing. Background technique [0002] The medical signs of pulmonary nodules are the basis for doctors to diagnose lung diseases. By analyzing various medical signs of lung CT images, it is convenient for doctors to judge the degree of benign and malignant nodules and make corresponding diagnostic decisions. However, doctors mainly diagnose diseases based on experience, and the diagnosis results are subject to some degree, so misdiagnosis and missed diagnosis often occur. Content-based medical image retrieval can help physicians quickly find similar lesion images from medical databases. The diagnostic schemes and lesion characteristics of these confirmed cases can provide references for the diagnosis of query lesions, thereby assisting physicians to make reliable diagnostic decisions. Contents of the invent...

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 Patents(China)
IPC IPC(8): G06K9/46G06T7/00G06N3/08G16H50/20
CPCG06T7/0014G16H50/20G06T2207/20084G06T2207/20081G06T2207/30064G06V10/462
Inventor 赵涓涓潘玲强梓林郝晓燕王华强彦
Owner TAIYUAN UNIV 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