Eureka AIR delivers breakthrough ideas for toughest innovation challenges, trusted by R&D personnel around the world.

Image classification method and device

A classification method and image technology, applied in the field of image processing, can solve problems such as inaccurate image classification

Active Publication Date: 2013-06-19
ALIBABA GRP HLDG LTD
View PDF2 Cites 18 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] The technical problem to be solved in this application is to provide an image classification method to solve the technical problem of inaccurate image classification in the prior art

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 classification method and device
  • Image classification method and device
  • Image classification method and device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0212] Corresponding to the method provided in Embodiment 2 of an image classification method of the present application, see Figure 6 , the present application also provides Embodiment 1 of an image classification device, which may include:

[0213] The feature extraction module 601 is used to extract the classified image features of an image to be classified.

[0214] The quantization determination module 602 is used to quantify each category image feature into a plurality of visual words in the visual dictionary according to the similarity relationship between each category image feature and the visual word in the pre-generated visual dictionary, and determine each category The similarity coefficients of image features and their quantized visual words respectively.

[0215] The division module 603 is configured to divide the image to be classified into multiple sub-images according to the image pyramid algorithm.

[0216] In order to enable the visual word histogram to r...

Embodiment 3

[0228] Corresponding to the method provided in Embodiment 3 of an image classification method of the present application, see Figure 7 , the present application also provides Embodiment 3 of an image classification device, which may include:

[0229] A feature extraction module 701, configured to extract a classified image feature of an image to be classified;

[0230] The model construction module 702 is configured to construct a sparse coding model of the classified image features and the pre-generated visual dictionary according to the similarity relationship between each classified image feature and the visual words in the pre-generated visual dictionary in a sparse coding manner.

[0231] Wherein, the sparse coding model is specifically:

[0232] arg C min Σ i = 1 N | | X i ...

Embodiment 4

[0244] Corresponding to the method provided in Embodiment 3 of an image classification method of the present application, see Figure 8 , the present application also provides Embodiment 4 of an image classification device, which may include:

[0245] A feature extraction module 801, configured to extract a classified image feature of an image to be classified;

[0246] The first calculation module 802 is configured to calculate the Euclidean distance between each classified image feature and the visual words in the visual dictionary according to the similarity relationship between each classified image feature and the visual words in the pre-generated visual dictionary.

[0247] Image features are expressed in vector form, for example, SIFT feature is a 128-dimensional vector. Visual words are obtained by clustering image features, which are also represented by vectors of the same dimension as image features. Among them, the Euclidean distance refers to the distance between...

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 provides an image classification method and a device. The method includes the following steps: extracting classification image characteristics of a to-be-classified image; quantizing each classification image characteristic into a plurality of vision words in a vision dictionary according to a similarity relation of the each classification image characteristic and the vision words in the preset vision dictionary, and determining similarity factors of the each classification image characteristic and the quantized vision words respectively; determining weight of the vision words to set a classified vision word histogram of the to-be-classified image according to the similarity factors of each vision word corresponding to different classification image characteristics in the vision dictionary; inputting the classified vision word histogram into an image classifier which is generated by training of a sample vision word histogram which is preset according to a large amount of sample images, and determining classes of the to-be-classified image according to output results. The image classification method and the device can improve accuracy of image classification and reduce classification errors.

Description

technical field [0001] The present application relates to the technical field of image processing, in particular to an image classification method and device. Background technique [0002] Image classification is an image processing technology that classifies images according to the characteristics reflected in different images to determine the category. With the rapid expansion of the number of images on the Internet, especially in the field of e-commerce, a large amount of image information is involved. Through image classification, the detection of prohibited items and the recommendation of similar products can be realized. Therefore, image classification technology has gradually become the focus of research. [0003] The existing image classification methods usually represent the image to be classified as a histogram of visual words to be classified according to the pre-generated visual dictionary, and then determine the category of the image to be classified by the imag...

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): G06K9/66G06V10/50
CPCG06V10/464G06V10/50G06V10/763G06F18/2321G06F18/214G06F18/22G06F18/24G06F2218/00G06F2218/12
Inventor 薛晖
Owner ALIBABA GRP HLDG LTD
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
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