Two-step sleepy driving prevention apparatus through recognizing operation, front face, eye, and mouth shape

Abstract

The present invention determines a period of a yawn, a speed and a period of a nod, and a speed and a period of a blink, of a user, using a driving determination unit which includes a GPS device and determines a driving state of a vehicle of a user and a sleepiness determination unit which receives an image and three-dimensional depth information photographed by a photographing unit including an infrared depth camera and an infrared camera which are installed at a location where the face of a user can be observed, in a first step determination, and when it is determined that there is a sleepiness prelude action, outputs a "preliminary warning". Thereafter, the present invention determines a closed state of a pupil, whether a gaze is headed to the front or not, and an angle of a head, in a following second step determination, and when it is determined that there is a sleepiness execution action, outputs a "main warning" through a warning unit. Thus, the present invention accurately determines only a sleepiness driving action except for a closed state of a pupil, a case where a gaze is not headed the front, and a case where a head is lowered, which does not result from sleepiness, thereby previously preventing a traffic accident, loss of life, and economic loss, which are generated due to sleepy driving.

Description

technical field [0001] The present invention relates to a device for preventing drowsiness through 2-step cognition of movements, facial eyes, and mouth shapes. It uses a GPS device to judge whether the user is driving forward, and only when the user is driving forward, begins to perceive the user's drowsiness state. In addition, the infrared depth sensor and infrared camera are used at the same time to recognize the driver's movements, face, eyes, and yawning mouth shape, so that it can be more accurately judged whether drowsy driving, and in order to rule out other reasons for non-drowsy driving. In order to accurately perceive drowsy driving, use a 2-step step-by-step method, first go through the first step of judging the precursor step of drowsiness and the second step of judging the execution step of drowsiness, and Drowsy driving is finally determined when one or more of the conditions in the first step is satisfied, and then one or more of the conditions in the second s...

AI Technical Summary

Problems solved by technology

[0003] However, as in the above-mentioned prior art, it also operates when the car is stopped. Therefore, when the driver closes his eyes or looks elsewhere in the stopped car, it is also judged to be drowsy driving, which causes inconvenience to the driver. In a backing car, when the driver is looking at the rear of the car or looking at the rearview mirror, it is also misjudged as drowsy driving for the reason of looking in other directions

[0004] And, the method of comparing and analyzing the vector template of the driver generated in the facial image reading section and the vector template of the driver stored in advance, as long as it is different from the vector template of the driver stored in advance, it is judged to be drowsy driving, It is difficult to distinguish whether it is drowsy driving or other reasons, and only because of a different reason than the pre-stored vector template, judging that it is drowsy driving can bring incorrect results

[0005] In addition, the conventional technology only judges the driver's eyeball area, or judges drowsy driving only by whether the eyeballs are opened or closed. , which greatly reduces accuracy by considering all cases where the user looks in other directions or closes or closes the eyeballs for reasons other than drowsiness as drowsiness

Moreover, the previous method for preventing drowsy driving judges the speed of the car as drowsy driving when it is above the average speed. This method has a fatal shortcoming, that is, when the driver's driving speed exceeds the average speed by a large amount due to other reasons other than drowsiness. , or when the surrounding environment forms obstacles and the movement or speed of the car cannot be recognized, it is impossible to judge whether it is drowsy or its accuracy is greatly reduced

Images