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

Image anomaly detection and positioning method and system based on discriminant learning

A positioning method and image anomaly technology, applied in neural learning methods, image enhancement, image analysis, etc., can solve the problems of limited improvement effect and inability to differentiate features, and achieve improved detection efficiency and accuracy, powerful performance and theory advantage effect

Pending Publication Date: 2022-07-29
UNIV OF JINAN
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, these methods cannot effectively differentiate the features, resulting in limited improvement

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 anomaly detection and positioning method and system based on discriminant learning
  • Image anomaly detection and positioning method and system based on discriminant learning
  • Image anomaly detection and positioning method and system based on discriminant learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0093] A method for detecting and locating image anomalies based on discriminant learning, the implementation steps of the method are as follows:

[0094] Step 1. Obtain image data to be detected, the data can be industrial images or medical images, and perform preprocessing operations on the images;

[0095] Step 11: Use the device to collect image data and define the original image as X∈R C×H×W ; where C represents the image dimension, H represents the height of the image, and W represents the width of the image;

[0096] Step 12, performing size scaling processing on the image data, the formula is as follows:

[0097]

[0098]

[0099] Among them, S(x,y) represents the corresponding pixel in the scaled image, (x,y) is the pixel coordinate, X represents the original image, X w and X h Indicates the width and height of the original image, S w and S h Indicates the width and height of the image after scaling;

[0100] Step 13, normalize the image data, the formula ...

Embodiment 2

[0137] An image anomaly detection and localization system based on discriminant learning, the system includes:

[0138] Image acquisition module for acquiring different types of images, including but not limited to industrial images and medical images;

[0139] The image preprocessing module is used to preprocess the obtained image;

[0140] Model training module, used to combine feature extraction network, feature preference selection and discriminant network model;

[0141] The model testing module is used to detect and locate the test set images;

[0142] Wherein, the image preprocessing module includes:

[0143] The data scaling module is used to scale the two-dimensional image data:

[0144]

[0145]

[0146] Among them, S(x,y) represents the corresponding pixel in the scaled image, X represents the original image, X w and X h Indicates the width and height of the original image, S w and S h Indicates the width and height of the image after scaling;

[0147]...

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 anomaly detection and positioning method based on discriminant learning, and the method comprises the following steps: carrying out the feature extraction of an original image through a network trained based on an external data set; carrying out feature selection by adopting a gradient preference method; and sending the obtained features into a discrimination network with center constraint to carry out anomaly detection. According to the method and the device, the abnormity existing in the image can be accurately detected and positioned under the condition that abnormal data is not needed. The method has strong performance and theoretical advantages in image anomaly detection and positioning. And the detection efficiency and accuracy are greatly improved.

Description

technical field [0001] The invention relates to the technical field of computer vision processing, and in particular provides an image abnormality detection and localization method and system based on discriminant learning. Background technique [0002] Image anomaly detection and localization is the detection and localization of abnormal areas in images, which has very high application value. The application of anomaly detection is very wide, mainly including medical image anomaly detection, industrial defect detection and so on. Especially in the industrial field, surface defects of industrial products such as metals, textiles, glass and wood boards have a negative impact on the safety and usability of the products. In the medical field, detecting and locating abnormalities in medical images can assist doctors in better diagnosis. Anomaly detection based on traditional manual detection and machine vision has disadvantages such as low detection efficiency, low accuracy, f...

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): G06V10/764G06V10/82G06V10/40G06V10/771G06K9/62G06T7/00G06T7/73G06T3/40G06T5/00G06N3/04G06N3/08
CPCG06T7/0002G06T7/73G06T3/4007G06N3/04G06N3/08G06T2207/20081G06T2207/20084G06T2207/20132G06T2207/20024G06F18/2113G06F18/2433G06T5/70
Inventor 牛四杰徐睦浩周雪莹高希占田京兰范雪张梦娇
Owner UNIV OF JINAN
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