Skin pathological image feature recognition method based on multi-example multi-label study

A technology of image features and recognition methods, applied in character and pattern recognition, instruments, computer parts, etc., can solve the problem of no skin pathological image feature recognition methods, and achieve the effect of improving effectiveness, efficiency and accuracy

Inactive Publication Date: 2013-09-25
GUANGDONG UNIV OF TECH
View PDF3 Cites 14 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

At present, there is no method specifically for feature recognition of skin pathology images.

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
  • Skin pathological image feature recognition method based on multi-example multi-label study
  • Skin pathological image feature recognition method based on multi-example multi-label study
  • Skin pathological image feature recognition method based on multi-example multi-label study

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0040] In this embodiment, the skin pathological picture feature recognition method based on multi-instance multi-label learning of the present invention includes the following steps:

[0041] 1) Image preprocessing;

[0042] 2) Regional feature extraction;

[0043] 3) Establish a skin pathology image recognition model based on multi-instance multi-label learning.

[0044]Because skin pathological images are mostly dyed and have special internal structures, the previous color space feature description methods suitable for skin surface images are not suitable for automatic recognition of pathological images.

[0045] The region division method of skin pathology image of the present invention is as follows:

[0046] The present invention uses the image represented by the RGB matrix as input, adopts "visually disconnected division method" in the present invention, regards each division as a part, and divides the picture into several parts that do not overlap visually. This way ...

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 skin pathological image feature recognition method based on multi-example multi-label study. According to the skin pathological image feature recognition method, an image is divided into a plurality of parts which are not overlapped in vision, feature extraction based on two-dimension wavelet transform is conducted on every part, and therefore a pathology image is expressed as a multi-example sample. Due to the facts that pathology image data already clear and definite in disease diagnosis by analysis are adopted, experience and specialized knowledge of doctors and researchers are abstractly extracted, and a machine learning model is used for expression, the model has the capacity of conducting sin pathology image feature recognition and taking the place of the doctors and the researchers, and diagnostic efficiency and accuracy are improved. The skin pathological image feature recognition method based on the multi-example multi-label study is skillful in design, good in performance, convenient to use and practical.

Description

technical field [0001] The invention is a feature recognition method of skin pathology pictures based on multi-instance and multi-label learning, which belongs to the transformation technology of the feature recognition method of skin pathology pictures based on multi-example and multi-label learning. Background technique [0002] In the process of diagnosing skin diseases, doctors need to carefully observe the changes on the skin surface of the patient's affected area. If it is recorded and saved as an image with photography technology, it will be a skin picture. For many skin diseases, it is impossible to make a correct diagnosis just by observing the patient's skin or skin pictures. Pathological pictures of skin tissue. Doctors can make a more accurate diagnosis through the analysis of skin pathology pictures. [0003] Since the pathological pictures reflect the histological features of the skin structure from the epidermis to the subcutaneous tissue, only doctors and l...

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): G06K9/62G06K9/46
Inventor 张钢苏向阳梁韵婷
Owner GUANGDONG 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