User concentration degree identification method and system based on hierarchical convolutional neural network

A convolutional neural network and recognition method technology, applied in the field of user focus recognition system based on hierarchical convolutional neural network, can solve problems such as rough classification of user focus, and achieve refined recommended programs, refined emotional classification, and results. Fine and accurate effects

Active Publication Date: 2017-09-22
WUXI YSTEN TECH
View PDF6 Cites 19 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Emotional classification or user concentration classification in existing smart TVs is too rough

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
  • User concentration degree identification method and system based on hierarchical convolutional neural network
  • User concentration degree identification method and system based on hierarchical convolutional neural network
  • User concentration degree identification method and system based on hierarchical convolutional neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0054] Such as figure 1 As shown, the first aspect of the present invention provides a method for identifying user concentration based on a hierarchical convolutional neural network, such as figure 1 shown, including the following steps:

[0055] S110 acquires the frontal image of the human face; the frontal image of the human face in the present invention can be acquired by receiving, or can be acquired by itself. When acquiring by itself, it is necessary to judge whether it is frontal. If it is not a frontal image, it can be discarded and re-acquired; in the present invention The frontal image of the face can be the frontal image directly obtained by the camera, or the image collected by the camera can be preprocessed to make the features clearer in order to facilitate feature extraction.

[0056] S120, according to the frontal image of the human face, utilize two local binary pattern (Local Binary Pattern LBP) operators of uniform patterns to calculate the feature encoding...

Embodiment 2

[0070] Such as Figure 4 As shown, the present invention provides a kind of user concentration recognition method based on hierarchical convolutional neural network, comprising the following steps:

[0071] Step.1 In the research of face pose estimation, the pose of the face is divided into three angles (pitch, yaw, roll), which represent the angles of up-down flip, left-right flip, and in-plane rotation. The present invention establishes the relationship between the key points of the human face in the two-dimensional plane and the rotation angle of the human face in the three-dimensional space by way of regression. Among them, the extraction of face key points adopts the SDM (Supervised Descent Method) algorithm. In the SDM algorithm, the NLS problem needs to be considered:

[0072] f(x)=min||h(x)-y|| 2

[0073] Here x is an optimization parameter, h is a nonlinear function, and y is a known variable. The following is an iterative formula based on gradient:

[0074]

...

Embodiment 3

[0110] Based on the method in Embodiment 1 and / or Embodiment 2, the present invention implements the above-mentioned method by programming controllers such as computers, MCUs, DSPs, and FPGAs, which not only includes devices such as hardware controllers, but also includes running and and / or stored computer code such as Figure 17 As shown, another aspect of the present invention also provides a user concentration recognition 100 based on a hierarchical convolutional neural network, including a face image acquisition device 110, a feature encoding map acquisition device 120, and a concentration acquisition device 130;

[0111] The human face image acquisition device 110 is used for a frontal image of a human face;

[0112] The feature encoding map acquisition device 120 is used to calculate the feature encoding map corresponding to the front face image of the human face by using two kinds of local binary pattern operators of uniform patterns according to the front image of the ...

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 user concentration degree identification method based on a hierarchical convolutional neural network. The method comprises the following steps that: obtaining the front side image of a face; according to the front side image of the face, utilizing two local binary pattern operators of an even pattern to calculate a feature coding graph corresponding to the front side image of the face; and according to the feature coding graph under two even patterns and the front side image of the face, adopting a GoogLeNet improved classifier to carry out classified processing to obtain the emotion of the user, and obtaining the user concentration degree according to the emotion. The invention also provides a user concentration degree identification system based on the hierarchical convolutional neural network. The user concentration degree result obtained by the invention is accurate and can be finely decomposed.

Description

technical field [0001] The present invention relates to the field of image processing, in particular, to a method for identifying user concentration based on a hierarchical convolutional neural network and a system for identifying user concentration based on a hierarchical convolutional neural network in an intelligent television system. Background technique [0002] With the popularity of machine learning, traditional human-computer interaction methods are gradually being eliminated. Nowadays, with the rapid development of smart TV systems, how to combine smart TV systems with machine learning to provide more convenient and humanized services is a problem worth exploring. Face recognition technology is a computer technology that uses the analysis and comparison of human face visual feature information for identity identification. It is a relatively popular technical research field and belongs to the field of biometric identification. Applying this technology to the TV syste...

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
Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/00G06K9/62G06N3/08
CPCG06N3/08G06V40/168G06V40/174G06F18/285
Inventor 黄飞侯立民邓卉李辉芳
Owner WUXI YSTEN TECH
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