BEV Lane Line Detection With Confidence-Based Sliding Windows
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing lane line detection technologies struggle to accurately identify stop lines in road images, affecting vehicle safety and driving experience.
Innovation Solution
A lane line detection method utilizing a vehicle-mounted device that converts foreground images to bird's-eye view (BEV) images, applies horizontal and vertical histograms, and uses sliding windows with confidence coefficients to accurately determine main lane lines and stop lines, enhancing recognition accuracy and robustness against weather conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional lane line detection methods are used, then the detection process is simple, but the accuracy of identifying stop lines and lane lines deteriorates
Solution Approach 1:
The detection method is segmented into distinct stages: image acquisition, coordinate system transformation to bird's-eye view, histogram calculation, sliding window detection, and lane line fitting. Each stage handles specific aspects of the detection task, improving overall accuracy while maintaining manageable complexity through modular processing
Solution Approach 2:
The patent transforms the detection problem from a two-dimensional image plane to a three-dimensional coordinate system by introducing height information and creating bird's-eye view images. This dimensional transformation enables more accurate identification of stop lines and lane lines by providing additional spatial context
2Reliability
If simple detection algorithms are used, then the processing speed is fast, but the reliability under adverse weather conditions deteriorates
Solution Approach 1:
The patent replaces traditional mechanical image processing methods with histogram-based statistical analysis and confidence coefficient calculations. This substitution enables more robust detection under varying weather conditions by using statistical properties of pixel distributions rather than simple thresholding or edge detection
Solution Approach 2:
The sliding window mechanism provides feedback by iteratively adjusting the detection region based on confidence coefficients and histogram peaks. This feedback loop continues until lane lines are successfully identified or the window reaches image boundaries, ensuring reliable detection while adapting to different weather conditions
3Measurement precision
If comprehensive image analysis is performed, then the detection accuracy is improved, but the computational complexity increases
Solution Approach 1:
The patent performs preliminary coordinate system transformation to bird's-eye view and calculates histograms before the actual lane line detection. These preliminary actions organize the image data in a way that simplifies subsequent detection steps and improve the efficiency of the overall processing pipeline
Solution Approach 2:
The sliding window approach applies different analysis methods to different regions of the image based on local characteristics. By focusing computational resources on regions with high confidence coefficients and relevant histogram peaks, the method achieves high detection accuracy without uniformly processing the entire image
Data Source
AI summary
A lane line detection method applied to a vehicle-mounted device is provided. In the method, the vehicle-mounted device converts a foreground image into a bird's-eye view (BEV) image, sets a peak of a horizontal histogram of the BEV image as a starting point for moving the sliding window, and generate a main lane line by fitting multiple non-zero pixels of the horizontal sliding window. The vehicle-mounted device calculates a confidence coefficient of each horizontal sliding window; and sets a previous one before the preset number of continuous horizontal sliding windows as a previous horizontal sliding window and determines an end point of the main lane line according to the previous horizontal sliding window when the confidence coefficients of preset number of continuous horizontal sliding windows are less than a predetermined threshold. The method improves accuracy of identifying lane lines.