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

Method for recognizing minority clothing images

A technology of minority and recognition methods, applied in character and pattern recognition, instruments, computer parts, etc., to achieve the effect of improving recognition efficiency, clear tone levels, and reducing over-learning problems

Inactive Publication Date: 2019-02-15
KUNMING UNIV OF SCI & TECH
View PDF3 Cites 11 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Few of the current known methods for minority clothing images

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
  • Method for recognizing minority clothing images
  • Method for recognizing minority clothing images
  • Method for recognizing minority clothing images

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0023] Embodiment 1: as Figure 1-6 As shown, a recognition method for minority clothing images, the specific steps are as follows:

[0024] Step1: Carry out human body detection on the image G to be recognized and the training image T, and use the k-poselet (k>1) deformable part model to detect each independent poselet, realize the overall and partial detection of the human body, and obtain the detected body parts to be recognized respectively Image G' and the detected training image T';

[0025] Step2 respectively extract five underlying features of the detected image to be recognized G' and the detected training image T' respectively color histogram, HOG, LBP, SIFT and edge operator, and obtain the image to be recognized after feature extraction G" and The training image T" after feature extraction;

[0026] Step3 Define the semantic attributes of ethnic minority clothing, mark the semantic attributes of the detected training image T', use the multi-task feature model to ...

Embodiment 2

[0031]Embodiment 2: In this embodiment, the image of ethnic minority clothing in Yunnan is taken as an example for illustration.

[0032] Step1, first input the minority clothing image G to be identified and input the training image T from the Yunnan minority clothing image library, use the weight vector ω=(M 0 ,...,M j ..., M k-1 , d 1 ,...,d j ... d k-1 ,b) describe each k-poselet, where, M j is the appearance template, d j is the spatial deformation model of the jth pose of k-poselet, b is the bias, and each k-poselet is described by a weight vector when detecting the model.

[0033] Then, k separate HOG templates are used to simulate the appearance model of each part, human detection is performed on each independent poselet, and keypoint prediction is performed from the average position of poselet positions and scales in the training data. Use average maximum precision (AMP) to measure whether a k-poselets set C achieves high precision and high coverage:

[0034] ...

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 relates to a method for recognizing minority clothing images, belonging to the field of computer vision, pattern recognition and image application. Firstly, the clothing images of minority nationalities to be recognized are inputted, and the images to be recognized and the training images in the clothing image database of minority nationalities are detected to obtain the images to berecognized and the training images respectively. Secondly, the color histogram, HOG, LBP, SIFT and edge of the detected image and training image are extracted to obtain the feature extracted image and training image. Then, the semantic attributes of minority clothing are defined, and the multi-task model is used to learn the different styles of minority clothing and to train the classifier model.Finally, through the trained classifier, the recognition of minority clothing images is realized and the recognition results are output. The identification method of the invention has high accuracy and efficiency.

Description

technical field [0001] The invention relates to a method for recognizing clothing images of ethnic minorities, and belongs to the fields of computer vision, pattern recognition and image applications. Background technique [0002] With the rapid growth of minority clothing image data, it is urgent to automatically analyze large-scale images by computer to extract relevant information that people can understand, so as to analyze, manage and identify these precious minority clothing images. Among the current known methods of clothing image recognition, most methods only focus on the category prediction of the underlying features and only for general clothing images. Chen et al. (<Describing clothing by semantic attributes>, 2012:609–623Heidelberg) proposed to extract the underlying features from body parts , including SIFT, color, and texture features, and then use the bag model and SVM to predict attributes and clothing categories. Shen et al. (<Unified structured l...

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
CPCG06F18/2411G06F18/214
Inventor 刘骊吴圣美付晓东黄青松刘利军
Owner KUNMING UNIV OF SCI & 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