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

An image recognition method of locality preserving projection

A technology for locally preserving projection and image recognition, applied in the fields of computer vision and image processing, to increase storage overhead, computational complexity, and improve robustness

Inactive Publication Date: 2019-09-03
BEIJING UNIV OF TECH
View PDF2 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The traditional LPP method is very sensitive to noise, because the square F norm amplifies the influence of noise on the algorithm

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 recognition method of locality preserving projection
  • An image recognition method of locality preserving projection
  • An image recognition method of locality preserving projection

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0012] Since the distance measurement of LPP is carried out in Euclidean space, if the data is projected onto the complex space represented by Euler, the influence of outliers will be reduced, thereby improving the robustness of the algorithm. At the same time, in the objective function of LPP, the distance measurement is based on the L2-norm, and one of its defects is that the outliers will be exaggerated under the action of the quadratic term. In this case, the learned projection matrix is ​​biased towards outliers away from the correct main direction. The traditional LPP method is very sensitive to noise, because the square F norm amplifies the influence of noise on the algorithm. Compared with the square F-norm, the method based on the F-norm is more robust to noise, which can weaken the negative impact of noise, improve the performance of the algorithm and enhance the robustness of the algorithm.

[0013] Such as figure 1 As shown, the image recognition method of this l...

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 present invention discloses an image recognition method of locality preserving projection which can reduce the effects of the abnormal values and the noise, thereby improving the robustness of themethod without increasing the storage overhead and the computational complexity of the image data. The method comprises the following steps of (1) establishing an analysis model for the input original image data X = [x1, x2,..., xN] through the local preserving projection based on Euler representation, wherein each image xi is a column vector, the size of each image xi is (2); adopting an iterative updating method to solve the model, and obtaining a projection matrix of the image; and (3) classifying the unknown images according to the projection matrix in the step (2).

Description

technical field [0001] The invention relates to the technical fields of computer vision and image processing, in particular to an image recognition method with local projection preservation, and is especially suitable for classification with abnormal values ​​in images. Background technique [0002] High-dimensional data is ubiquitous in modern computer vision and image processing research. However, high-dimensional data will not only increase storage overhead and computational complexity, but also reduce the effectiveness of algorithms in practical applications. High-dimensional data are often distributed in low-dimensional subspaces or low-dimensional structures of manifolds. Therefore, finding the mapping relationship between high-dimensional data and low-dimensional space has become an important issue for image classification. Algorithms for data dimensionality reduction have made extensive progress in recent decades. [0003] Locality Preserving Projection (LPP) is a...

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): G06K9/62
CPCG06F18/213G06F18/24
Inventor 孙艳丰龙天航胡永利
Owner BEIJING UNIV 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