Detection system of road lane line and method for same

Abstract

The invention discloses a detection system of a road lane line and a method for the same. The system comprises image capture equipment used to capture image frame data in front of a vehicle and transmit the image frame data to a processing device. Multiple feature values converted by one or two road lane lines are gradually obtained from the part under the image frame data to the upper position; the feature values are processed by the processing device, so one or two predicted road lane lines can be generated in the image frame data; then, the predicted road lane lines are used to dynamically predict a complete road lane line in the image frame data, so that a road lane line prediction model is formed; the processing device transmits the image frame data and the road lane line prediction model to display equipment for display; and when the vehicle deviates, the display equipment will send alarms. According to the invention, identification effect of fuzzy, single and far road lane lines can be enhanced.

Description

technical field [0001] The present invention relates to a lane line detection system and method, in particular to a road lane line detection system and method for predicting lane lines on a road by using an algorithm and image processing. Background technique [0002] With the popularization of vehicles, the vehicles running on the road are easy to gather into a huge traffic flow, and many vehicles form a potential crisis on the road at the same time. According to statistics, the collision accidents caused by the driver's vehicle deviation from the lane for various reasons account for 51% of all traffic accidents, which are frequent and serious traffic accidents. [0003] Therefore, many technologies and products related to driving vehicles have been gradually developed, such as Lane Departure Warning (LDW), Lane Keeping Support (LKS) and Lane Following System (Lane Following). System, LFS), all of which are used to assist driving safety and prevent accidents. These systems...

AI Technical Summary

Problems solved by technology

In the general lane line recognition method, the main method is to use the car camera to capture the image frame, and use the processor to fit the feature points of all the lane lines on the image frame. This method is prone to noise points and unclear remote feature points. Influence, for example, one of the two lane lines on the actual road is unclear due to damage or occlusion, which makes the processor unable to effectively fit the lane line fitting model, or the remote lane line is far away, in the The image screen is relatively unclear, and it is not easy to identify feature points, which leads to poor fitting effect

Images