A system and method for identifying directional traffic markings in an urban environment

Abstract

The invention relates to a guide traffic marking identification system and a guide traffic marking identification method in a municipal environment, in order to realize that a remotely piloted vehicle actively cognizes a road surface guide identifier. The guide traffic marking identification system provided by the invention comprises an image acquisition unit, a guide traffic marking extracting unit, a database unit, a guide traffic marking identification unit and a vehicle-mounted control unit, wherein the image acquisition unit acquires image information shot on a road; the guide traffic marking extracting unit extracts guide traffic marking identifiers from the images acquired by the image acquisition unit; the database unit stores identifiers for guiding the traffic markings; the guide traffic marking patterns extracted by the guide traffic marking identification unit are compared with the guide traffic marking identifiers of the database unit, and the identifying results of the patterns are transmitted to the vehicle-mounted control unit; and the vehicle-mounted control unit sends a vehicle driving instruction according to the identifying result of the guide traffic marking identifying unit. According to the system and the method, the unmanned vehicle can initiatively cognize the road surface identifiers, feature dimensions can be reduced by an independent two-dimension component analysis method, and the categories of traffic markings can be classified by a nearest classifier; and the feature dimension obtained by the method is low, and the classifying accuracy is high.

Description

technical field [0001] The invention belongs to an image recognition system, and in particular relates to a traffic marking that can recognize signs on urban roads, and is used to guide unmanned vehicles to perceive the road environment, make intelligent decisions on driving routes, and comply with traffic laws and regulations. Background technique [0002] At present, in the field of unmanned vehicles and active safety, the requirements for the environmental information perception of vehicles are increasing. Not only must the information of various sensors on the vehicle be used to obtain information about obstacles around the vehicle that affect driving, but also according to the environment in which the vehicle is located. , intelligent decision-making routes and compliance with traffic laws and regulations. [0003] At the intersections in the urban road environment, there are generally traffic markings composed of various lines, arrows, characters, raised road signs and...

AI Technical Summary

Problems solved by technology

[0004] However, at present, there are few studies on the traffic marking recognition system on the ground, especially in the application of driverless vehicles and active safety.

Images