Autonomous driving assistance glasses that assist in autonomous driving

Abstract

A method for providing safe-driving information via eyeglasses of a driver of a vehicle is provided. The method includes steps of: a safe-driving information analyzing device, (a) if a visual-dependent driving image, corresponding to a perspective of the driver, from a camera on the eyeglasses, acceleration information and gyroscope information from sensors are acquired, inputting the visual-dependent driving image into a convolution network to generate a feature map, and inputting the feature map into a detection network, a segmentation network, and a recognition network, to allow the detection network to detect an object, the segmentation network to detect lanes, and the recognition network to detect driving environment, inputting the acceleration information and the gyroscope information into a recurrent network to generate status information on the driver and (b) notifying the driver of information on a probability of a collision, lane departure information, and the driving environment, and giving a warning.

Description

technical field [0001] The present invention relates to a pair of auxiliary glasses worn by a driver, and in more detail, relates to a pair of auxiliary glasses that acquire information from a camera mounted on the auxiliary glasses worn by the driver by analyzing based on deep learning. Driving safety information is provided based on the visually dependent driving image corresponding to the driver's perspective. Background technique [0002] Today's vehicles provide various functions in combination with IT technology, and various advanced driver assistance systems (Advanced Driver Assist System, ADAS) are being developed to enhance the driving stability of the vehicle and improve user convenience. [0003] At this time, in order to provide various information for autonomous driving, ADAS including pedestrian and vehicle recognition systems, road surface recognition systems, lane recognition systems, collision avoidance systems and Lane departure warning system, etc. [00...

AI Technical Summary

Problems solved by technology

[0004] However, the cost increases due to the use of various sensors and cameras for recognizing the driving environment of the car and the state of the driver in such existing ADAS

[0005] In addition, in the existing ADAS, the sensors and cameras need to be installed in the optimal positions for obtaining relevant information, but because accurate information cannot be obtained due to errors in position and sensing direction, etc., it is impossible to The problem of providing accurate information for driver assistance

Images