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

An Expectation-Maximization Parameter Estimation Method for Gaussian Mixture Models

A Gaussian mixture model and expectation maximization technique, applied in the field of information processing, which can solve problems such as fuzzy color features

Active Publication Date: 2012-02-22
BEIJING INSTITUTE OF TECHNOLOGYGY
View PDF3 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the above step ① has already reduced the sample data, and the parameters input into GMM are fuzzy color features. The current EM algorithm cannot be directly applied to this situation. An improved EM algorithm needs to be provided to complete the parameter estimation process

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

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0033] The present invention provides a method for estimating EM parameters in a Gaussian mixture model, which is applied to the system based on fuzzy and Gaussian mixture models described in the background art.

[0034] The specific process of applying the EM parameter estimation method of the present invention to the above-mentioned fuzzy and Gaussian mixture model to realize skin color pixel recognition will be described in detail below.

[0035] figure 1 It is a schematic diagram of the skin color pixel recognition scheme based on fuzzy and Gaussian mixture model in the present invention. The scheme includes three parts: pixel feature extraction process, GMM construction process and pixel feature recognition process. The above three processes are described separately below.

[0036] ●Pixel feature extraction process

[0037] The present invention adopts the brightness and color difference YCbCr color space commonly used in skin color pixel recognition. The main reason f...

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 method for estimating parameters of expectation maximization in a Gaussian mixture model. The method improves the specific calculation process of the EM algorithm on the basis of sample clustering, and provides the estimation of three parameters of w, μ and ∑ method, making it suitable for sample reduction and the input parameters of GMM are fuzzy color features. Due to the reduction in the number of samples, the parameter estimation process will not introduce unnecessary noise, thereby improving the accuracy of parameter estimation; due to the reduction in the number of samples, the improved EM algorithm is also improved in terms of storage capacity, calculation capacity, and running time. , which is beneficial to improve the efficiency of parameter estimation.

Description

technical field [0001] The invention relates to the technical field of information processing, in particular to a parameter estimation method for expectation maximization in a Gaussian mixture model. Background technique [0002] Skin color pixel recognition in images is commonly used in image recognition technologies related to the human body, such as adult image recognition, face recognition, and gesture recognition. Due to the diversity of image types and content, including various races under different lighting, background and makeup conditions, it is difficult to realize skin color pixel recognition technology. [0003] Skin color pixels are a kind of underlying feature, and its recognition results are extremely important for subsequent processing. In different application environments, subsequent processing has different requirements for the accuracy and speed of skin color pixel recognition. Therefore, in order to obtain a more general The skin color pixel recognitio...

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): G06K9/00G06T7/00
Inventor 胡昌振王潇茵姚淑萍
Owner BEIJING INSTITUTE OF TECHNOLOGYGY
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