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

A real-time face detection method for day and night drivers based on rgb-i

A real-time detection and driver technology, applied in the direction of instrument, calculation, character and pattern recognition, etc., can solve the problem of not reaching real-time, and achieve the effect of improving time, speed and accuracy

Active Publication Date: 2019-12-31
JIANGSU UNIV
View PDF9 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] At present, there are many literatures on face detection, most of which mainly analyze multi-angles in complex backgrounds and cannot reach real-time. However, there are still few studies on driver's face detection.

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
  • A real-time face detection method for day and night drivers based on rgb-i
  • A real-time face detection method for day and night drivers based on rgb-i
  • A real-time face detection method for day and night drivers based on rgb-i

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0034] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments.

[0035] figure 1 The general idea of ​​the present invention is given, use the RGB-Infrared camera to obtain the driver's image of day and night, in the daytime, obtain the driver's RGB image in real time, slide the window with 40*40 for each frame of image, the step size is 2, The image obtained by the sliding window is input to the trained R40-Convnet model to detect whether it is a human face. If it is a human face, the size of the image is adjusted to 28*28 and then input to the next trained R28-Convnet model. For the face, save the position of the window at the moment, otherwise it will not be saved, and then continue to slide the window until the complete test picture is detected, then scale the picture at a ratio of 0.8, repeat the above steps until the picture is smaller than 40*40, then When , there are many overlapping windows at 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 discloses a real-time face detection method for drivers driving day and night based on RGB-I, which includes the steps of model training and driver face recognition; the step of model training includes: Processing: Divide the driver’s face image into two groups with sizes of 40*40 and 28*28, and then grayscale; S2, establish the R-Convnet model based on RGB image training and the I-Convnet model based on Infrared image training , the R-Convnet model is divided into R40-Convnet model and R28-Convnet model, and the I-Convnet model is divided into I40-Convnet model and I28-Convnet model. The four Convnet models are obtained by removing the fully connected layer on the basis of CNN; driving The steps of driver’s face recognition include: S3, use RGB camera and Infrared camera to collect driver’s images in daytime and nighttime respectively; S4, use trained cascaded R40‑Convnet model and R28‑Convnet model for recognition in daytime, and use it at night The trained cascaded I40‑Convnet model and I28‑Convnet model are used for recognition. The invention improves the time and accuracy of driver's face detection at night or in bad driving environment.

Description

technical field [0001] The invention relates to a method for real-time detection of a driver's face, in particular to a real-time detection method for a driver's face in day and night driving based on RGB-I. Background technique [0002] In recent years, people's living standards have been improving day by day, the number of private cars has been increasing, and the incidence of traffic accidents has also been increasing. Vehicle and safety issues have become a hot spot of social concern. Safety issues include driver fatigue detection, drunk driving, driver emotion detection, driver behavior detection, etc. In driver fatigue detection and expression detection, the first step is to detect the position of the face (face detection). [0003] Face detection is the process of determining the position, size, and pose of all faces (if present) in an input image. The problem of face detection originally originated from face detection. The research on face detection can be traced ba...

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): G06K9/00
CPCG06V40/161G06V40/172
Inventor 毛启容刘鹏高利剑刘峰陈龙詹永照
Owner JIANGSU UNIV