Fault recognition method for fan blade images based on nuclear norm regularized low-rank coding

A technology for fan blade and fault identification, applied in the field of image identification, can solve problems such as inaccurate solution of representation coefficients, low-rank information not being well utilized, etc., to achieve the effect of improving identification accuracy

Active Publication Date: 2019-04-12
NANJING UNIV OF POSTS & TELECOMM
View PDF2 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, those training samples with the smallest residuals may not be from the same class as the test samples, so that the non-zero elements in the solved representation coefficients do not only correspond to the samples of the same type, resulting in the inaccurate solution of the representation coefficients, and at the same time, the same class of training samples The low-rank information of the sample is not well utilized

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
  • Fault recognition method for fan blade images based on nuclear norm regularized low-rank coding
  • Fault recognition method for fan blade images based on nuclear norm regularized low-rank coding
  • Fault recognition method for fan blade images based on nuclear norm regularized low-rank coding

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0022] Embodiments of the present invention are described in detail below, examples of the embodiments are shown in the accompanying drawings, and the embodiments described with reference to the drawings are exemplary and are only used to explain the present invention, and cannot be construed as limitations of the present invention .

[0023] The present invention provides a fan blade image fault recognition method based on nuclear norm canonical low-rank coding, the specific process is as follows figure 1 shown.

[0024] (1) Obtain training sample set

[0025] Assuming that the size of the image is w×h, the training samples come from c image classes, and the training samples of the i (i={1,2,…,c}) class are expressed as (in, j={1,2,...,n i}). The training sample set can be expressed as in, Indicates the number of all training samples.

[0026] For training sample B ij , and normalize it modulo 1:

[0027]

[0028] Among them, B ij (p,q) is the training sampl...

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 fan blade image fault recognition method based on nuclear norm regular low-rank coding. The method first obtains a training sample set of fan blade fault images, and then uses the low-rank coding method based on nuclear norm regular to obtain samples to be identified in The linear representation coefficient on the training sample set, and finally by calculating the representation residual of the sample to be recognized on each class, the class label of the sample to be recognized is calculated according to the residual. The method of the invention not only maintains the structural information of the image, but also integrates the low-rank features of the same sample itself into it, and uses the sparse feature information to obtain the representation coefficient of the sample to be recognized, thereby improving the recognition accuracy.

Description

technical field [0001] The invention relates to an image recognition method, in particular to a fan blade image fault recognition method based on nuclear norm regular low-rank coding, which belongs to the technical field of image recognition. Background technique [0002] Image recognition is generally divided into four steps: image preprocessing, feature extraction, feature expression, and classification. In order to better track and identify the operating status of wind turbine blades, it is often necessary to take some images of wind turbine blades. Traditional image recognition algorithms assume that the input image is of good quality. However, in real life, since the target image is often far away from the camera equipment, and is affected by factors such as changes in lighting conditions, motion blur of the target image, and noise of the device itself, the acquired image has low resolution and large noise. Feature details are also extremely limited. In this case, th...

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): G06K9/62
CPCG06F18/214
Inventor 高广谓岳东荆晓远吴松松邓松
Owner NANJING UNIV OF POSTS & TELECOMM
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