Unlock instant, AI-driven research and patent intelligence for your innovation.

Customization classifier-based eye state identification method

A technology of eye state and recognition method, which is applied in the field of driver fatigue detection and can solve problems such as difficult classifier discrimination

Active Publication Date: 2012-04-25
HOPE CLEAN ENERGY (GRP) CO LTD
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

This is due to the great differences in the opening and closing of eyes of each person, as well as habits such as whether to wear glasses, it is difficult to use a general classifier to distinguish

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
  • Customization classifier-based eye state identification method
  • Customization classifier-based eye state identification method
  • Customization classifier-based eye state identification method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0037] An eye state recognition method based on a user-defined classifier, such as figure 1 shown, including the following steps:

[0038] Step 1: Establish face image database A. The face database A includes two sub-databases A1 and A2, wherein one sub-database A1 is composed of grayscale images of faces of different individuals, without glasses, and the front face of people other than users, and the other sub-database A2 is composed of images of faces other than users. It is composed of gray-scale images of faces outside, different individuals, wearing glasses, and front faces. The distance between the center points of the two eyes of the face grayscale image in the face database A is not less than 48 pixel units, and the number of face grayscale images in the eyes-open state and the eye-closed state is basically the same.

[0039] Step 2: Establish user face image database B. The user's face image database B includes two sub-bases B1 and B2, wherein one sub-base B1 is c...

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 belongs to the technical field of image processing and mode identification and is suitable for driver fatigue detection. The method comprises the following steps of: establishing a human face image library and a user face image library, calculating eye images of each image and mixing the two libraries according to different proportions; calculating the haar-like characteristic vector of each image in the mixed eye image library and constructing a strong classifier by using an AdaBoost method; randomly selecting a plurality of eye images in user face image library, judging the constructed strong classifier, and selecting the strong classifier with the highest identification accuracy as the eye state identification classifier used when the user drives. Through the method, different classifiers are used for different users by a method of mixing the user data and the human face library data according to the customization concept, so that the identification accuracy of the classifier is improved and the identification risks are reduced. The invention further provides two different classifiers for users wearing or not wearing glasses, and the eye state identification is more flexible.

Description

technical field [0001] The invention belongs to the technical field of image processing and pattern recognition, and relates to driver fatigue detection technology. Background technique [0002] At present, traffic accidents cause tens of thousands of vehicle collisions and major casualties every year. According to incomplete statistics, the number of deaths caused by road traffic accidents in the world exceeds 600,000, of which at least 100,000 traffic accidents are caused by driver fatigue. 100,000 cases, the direct economic loss amounted to 12.5 billion US dollars. Driver fatigue driving has become the main hidden danger of traffic accidents just like drunk driving. With the development of computer technology, researchers from various countries have begun to study the detection methods of fatigue driving in depth. In 1998, the US Federal Highway Administration test confirmed that PERCLOS (percentage of human eye closure per unit time) has a significant relationship with ...

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 Patents(China)
IPC IPC(8): G06K9/66A61B5/18
Inventor 马争解梅孙睿
Owner HOPE CLEAN ENERGY (GRP) CO LTD