Multi-group image classification method based on characteristic expansion and fuzzy support vector machine

A technology of fuzzy support vector and classification method, applied in the field of image processing, can solve the problems of low classification accuracy and the inability of multi-group image classification method to effectively extract the essential features of images, so as to improve the classification accuracy and avoid the disaster of dimensionality.

Active Publication Date: 2012-08-01
哈尔滨工业大学高新技术开发总公司
View PDF2 Cites 8 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to solve the problem that the existing multi-group image classification method cannot effectively extract the essential f...

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
  • Multi-group image classification method based on characteristic expansion and fuzzy support vector machine
  • Multi-group image classification method based on characteristic expansion and fuzzy support vector machine
  • Multi-group image classification method based on characteristic expansion and fuzzy support vector machine

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach 1

[0032] Specific implementation mode one: the following combination figure 1 Describe this embodiment, the multi-group image classification method based on feature extension and fuzzy support vector machine described in this embodiment, the method comprises the following steps:

[0033] Step 1. Initialize the given number of bands as I 0 , a multi-group image of size P×Q

[0034] IM j (p, q), j=1, 2, ..., I 0 , p=1, 2,..., P, q=1, 2,..., Q,

[0035] Remove multi-group image IM j (p, q) are severely polluted by noise and other bands that cannot be used, and the remaining I effective bands are reordered to obtain multi-group images to be expanded

[0036] IM i (p, q), i=1, 2, ..., 1,

[0037] where I 0 , I, P and Q are natural numbers;

[0038] Step 2, successively to I multi-group image IM to be extended i (p, q) perform two-dimensional empirical mode decomposition to obtain the I group of two-dimensional eigenmode functions

[0039] BIMF ...

specific Embodiment approach 2

[0052] Specific implementation mode two: the following combination figure 2 Describe this embodiment mode, this embodiment mode will be further described to embodiment mode one, in step 2 sequentially to 1 multi-packet image IM to be extended i (p, q) perform two-dimensional empirical mode decomposition to obtain the I group of two-dimensional eigenmode functions The process is:

[0053] Step A. Initialize r 1 =IM i (p,q); u=1; v=0; SD=1000; h u，v = r 1 ; c u = r 1 ,

[0054] Among them, the to-be-expanded multi-group image of the i-th band to be processed is IM i (p,q),

[0055] r 1 is the residual after the first two-dimensional empirical mode decomposition,

[0056] SD is the termination iteration threshold,

[0057] h u，v is the residual function after the vth screening in the uth two-dimensional empirical mode decomposition,

[0058] c u is the two-dimensional eigenmode function

[0059] Step B, let v=v+1; h u，(v-1) = r u , and by comparing with adja...

specific Embodiment approach 3

[0074] Specific implementation mode three: this implementation mode will further explain implementation mode one, and in step three, all two-dimensional eigenmode functions Organic combination, the process of obtaining the extended feature FBIMF is:

[0075] The i=1, 2, ..., the two-dimensional intrinsic mode function of the I band are processed in sequence, and the extended feature FBIMF corresponding to the two-dimensional intrinsic mode function of each band is obtained, and the two-dimensional intrinsic mode function of each band is The method of obtaining the extended characteristic FBIMF of the intrinsic mode function is the same, as follows: in the two-dimensional intrinsic mode function of the band i = 1, 2, ..., U i For a BIMF, if u i is an odd number, then the link directly to tail if u i is an even number, then the Flip left and right and connect to Tail, in turn, until all U of the two-dimensional intrinsic mode function of the band i All BIMFs are pro...

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 multi-group image classification method based on characteristic expansion and fuzzy support vector machines, belonging to the field of image processing. The invention aims to solve the problems that the existing multi-group image classification method can not effectively extract the substantive characteristics of an image and the classification precision is relatively low. The method comprises the following steps of: firstly removing the wave bands which can not be used due to the serious pollution from noise and the like, and performing two-dimensional empirical mode decomposition on the remaining wave bands to obtain some two-dimensional intrinsic mode functions; organically combining the two-dimensional intrinsic mode functions, and expanding into the characteristics of multi-group images; and finally, classifying by a fuzzy support vector machine serving as a classifier. The method disclosed by the invention gives full play to the advantage that the two-dimensional empirical mode decomposition can adaptively extract the substantive characteristics of a complex image, and effectively obtains the characteristics of multi-group images; and moreover, by adopting the fuzzy support vector machine as a classifier which integrates the advantages of a support vector machine and a fuzzy function, the classification precision is improved.

Description

technical field [0001] The invention relates to a multi-group image classification method based on feature extension and fuzzy support vector machine, belonging to the field of image processing. Background technique [0002] Multi-group images are composed of multi-band images with high correlation. There are a large number of physical prototypes in the fields of environmental monitoring, earth survey, medical diagnosis, radar detection and military investigation, such as: sea level fluctuation images, hyperspectral images and medical ultrasound images, etc. Multi-group images are generally continuously observed for the same area or location or spectroscopically observed by an imaging spectrometer, and often contain hundreds or thousands of bands with high correlation, and each pixel corresponds to a characteristic curve covering each band. It embodies the multi-resolution information of the observed object, and also contains a large amount of redundant information. [000...

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
Inventor 沈毅贺智张淼
Owner 哈尔滨工业大学高新技术开发总公司
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