Traffic light intersection blind guiding method and blind guiding device based on machine vision

Abstract

The invention discloses a traffic light intersection blind guiding method and blind guiding device based on machine vision, and the blind guiding method comprises the steps: collecting the environment image and zebra crossing image information of a traffic light intersection, extracting the color of a traffic light and countdown information, judging whether the traffic light can pass through the intersection, extracting the zebra crossing information, and judging whether a user deviates from a correct route or not, and prompting the user whether it is suitable to pass the intersection. The blind guiding device comprises a mirror frame, and a power supply module, a central processing module, a voice broadcast module, an image acquisition module and a wireless communication module which are arranged on the mirror frame; the device can provide accurate guidance for the blind needing to pass through the traffic light intersection, guarantees the personal safety of the blind in daily traffic, and is high in reliability and precision, simple in structure and convenient to carry.

Description

technical field [0001] The invention belongs to the technical field of machine learning, and in particular relates to a blind-guiding method and a blind-guiding device at a traffic light intersection based on machine vision. Background technique [0002] Along with the raising of national economy level, urban traffic and automobile industry develop vigorously, make a large number of crossroads add in many places, have caused great inconvenience for this group of blind people, cause many blind people to stay away from traffic light intersections and even avoid going out, to Their life and work have caused great inconvenience. [0003] Blind people mainly include two categories, one is total blindness, which refers to people who lose their response to light and cannot see at all, and the other is semi-blindness, which refers to people whose vision is less than 60% of normal vision. According to statistics, as of 2018, there are blind people in every 82 people in our country. ...

AI Technical Summary

Problems solved by technology

[0004] But there is great safety in said method, mainly comprises the following several aspects: (one) the environment of traffic lights is more complicated with respect to other environments, and sound is noisy, is not convenient for blind people to listen to the engine sound of parked car; (two) ) There are often uncivilized phenomena such as running through traffic lights at traffic lights. If you rely on the sound of pedestrians’ footsteps to judge whether you can pass, it is very easy to make a wrong judgment, which will lead to safety accidents; (3) currently only some major intersections in a few cities use traffic lights (4) Relying on the above method, even if the current status of traffic lights is obtained, the remaining time of red and green lights cannot be judged, which has great potential safety hazards; (5) There is no paved blind road on the zebra crossing, and blind people can easily deviate from the direction when passing, and even go to the dangerous area

Images