Lane Line Detection Using Direction Density Grouping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional lane line detection methods in automatic driving are susceptible to image quality and noise, requiring cumbersome post-processing and lacking robustness.
Innovation Solution
A lane line processing method that calculates direction densities of individual lane line points, groups them accurately, and uses polynomial fitting to obtain robust lane line representations, incorporating a deep learning model for initial classification and scaling.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional image processing methods are used for lane line detection, then the detection process can be completed, but the results are susceptible to noise points and image quality, requiring cumbersome post-processing
Solution Approach 1:
The patent replaces conventional image processing methods (edge detection, clustering, Hough transform) with a deep learning-based approach. The neural network directly outputs lane line information from the input image, substituting the mechanical/image processing pipeline with an intelligent system that learns features automatically, thereby eliminating the need for cumbersome post-processing while improving robustness against noise and varying image quality
Solution Approach 2:
The patent transforms the lane line detection problem by changing the output parameters of the deep learning model. Instead of outputting raw pixel coordinates requiring post-processing, the model outputs refined lane line parameters (position, curvature, orientation) directly, changing the parameter representation from raw image data to meaningful geometric descriptors that are immediately usable
2Measurement precision
If conventional image processing methods are used for lane line detection, then the detection can proceed, but the accuracy is affected by image quality and noise
Solution Approach 1:
The patent replaces conventional image processing methods (edge detection, clustering, Hough transform) with a deep learning-based approach. The neural network directly outputs lane line information from the input image, substituting the mechanical/image processing pipeline with an intelligent system that learns features automatically, thereby eliminating the need for cumbersome post-processing while improving robustness against noise and varying image quality
Solution Approach 2:
The deep learning model performs preliminary feature extraction and noise filtering during the forward propagation process. By pre-training on large datasets with various noise conditions, the model learns to robustly identify lane lines before the actual detection occurs, performing the noise-resistant feature extraction in advance rather than requiring post-processing cleanup
Data Source
Figure 1
Figure 2
Figure 3~4
AI summary
A lane line processing method and device are provided. The method includes: obtaining distances between lane line points in a first image; determining direction densities of the individual lane line points by using the distances between the lane line points; dividing lane line points into groups corresponding to lane lines by using the direction densities of the individual lane line points; and obtaining representation information of the lane lines corresponding to the groups by using the lane line points in the groups. By using the direction densities of the individual lane line points, lane line points in an image are divided into groups, the obtained groups are then more accurate, and thus the lane lines obtained by fitting are accurate, are not susceptible to image quality, and have high robustness.