Eye portion motion capture method based on projection integration and iris recognition

Abstract

The present invention discloses an eye portion motion capture method based on projection integration and iris recognition. The method comprises the steps of: S1, face image obtaining; S2, image preprocessing, S3, projection integration calculation, S4, human eye area location, S5, location of an iris area, and S6, capture of eye portion motion. The method provided by the invention employs gray-level projection integration to combine face structure features with the gray-level projection integration, gray scale change of the human eye area in horizontal and vertical directions is calculated toobtain location of the human eye area, an iris physical structure features are employed to locate the center position of the iris in an eye area image determined by the eye position, and finally, a parameterization template is used to achieve identification of location of motion of upper and lower eyelids in the eye local area image; and moreover, the eye portion motion capture method based on projection integration and iris recognition overcomes the imaging difference of the eye images in different illumination conditions, improves the adaptability of the detection system for the illuminationand achieves the all-weather and robust driver's eye portion motion captures.

Description

technical field [0001] The invention belongs to the field of vision detection, and relates to an eye motion capture method based on projection integral and iris recognition. Background technique [0002] Fatigue driving refers to that due to the monotony of the driving environment or long-term, super-intensive driving, the driver’s physical and psychological functions decline due to excessive energy consumption, resulting in a decline in response level and control efficiency, which in turn leads to a reduction in driving skills and affects Phenomenon of normal driving. When the driver is fatigued, his ability to perceive the traffic environment, his ability to judge danger, and his ability to control the vehicle all decline to varying degrees, which can easily lead to traffic accidents. [0003] The mainstream detection method for fatigue driving is to evaluate the driver's mental state with the driver's participation. The most commonly used method is the fatigue detection ...

AI Technical Summary

Problems solved by technology

[0006] The purpose of the present invention is to overcome the deficiencies of the prior art, especially to solve the problem of random illumination in the actual driving process, driver's body movements, blurred and dim imaging under night working conditions, and dark glasses working conditions in the existing facial video fatigue detection method. Eye image has invisible effects, causing eye movement to fail

Images